GEMS M&E Handbook

Version 1.0 | December 2012

GEMS RESULTS
MEASUREMENT
HANDBOOK

This Handbook has been prepared by ITAD Ltd on behalf of the GEMS
programme in Nigeria. For further information, please contact Tim Ruffer,
Associate Director at ITAD: tim.ruffer@itad.com

www.gemsnigeria.com
www.itad.com

GEMS Results Measurement Handbook

CONTENTS
Foreword ii
Abbreviations iii
1.

Handbook purpose

1

2.

GEMS strategic framework

3

3.

Measurement approach

8

3.1.

9

3.2.

A 3-Step approach to measuring and attributing
change
Six stages of measurement

11

4.

Defining indicators of change

13

5.

Selecting measurement approaches

21

5.1.
5.2.
5.3.

Applying a mixed methods approach
Selecting measurement methods
Aggregating results

22
23
26

6.

Estimating attributable change

27

7.

Data sources and collection methodologies

30

7.1.
7.2.
7.3.
7.4.

Data sources
Data collection strategy
Survey guidance
Measurement frequency

31
31
32
32

Annexes

33

Annex 1: GEMS logframe
Annex 2: Glossary of terms
Annex 3: References and further reading
Annex 4: Nigeria household composition by zone (2011)
Annex 5: Intervention-specific guidance
Annex 6: Categorization of GEMS interventions

34
36
37
38
39
60

i

ii


FOREWORD
The Growth and Employment in
States (GEMS) project in Nigeria
is responding to a number of
fundamental monitoring, evaluation
and reporting challenges. First and
foremost, it is a large and complex
project, implemented by different
companies, across different states and
sectors, with a very wide variety of
interventions. Second, it makes use of
the Making Markets Work for the Poor
approach to economic development.
M4P has been challenging the talents
and persistence of M&E specialists
and economists for some years, as
they have tried to resolve issues
of attribution, displacement and
deadweight within fluid and massive
market systems. Finally, the project
is delivered in a period of global
development practice dominated by
demands for effective communication
of results and value for money.
This Handbook responds directly
to these challenges, and as a result,
provides three main contributions
to GEMS, and potentially to other
M4P projects beyond Nigeria. First,
it provides a set of definitions for key
terms (the poor, ‘income increase’,
employment etc) whose meanings
we might think we understand, but
which quickly unravel when two sets
of results using different methods and
measurements (and, crucially, different
M&E specialists) are compared. The

unusual complexity of GEMS, notably
its use of three main service providers,
has forced the work to be particularly
detailed and thoughtful, and should
inform work in other, similar projects.
In addition to definitions, it provides
a set of measurement standards
and guidelines for each category of
intervention in the project, whether
focused on skills, introduction of new
products and services, advocacy,
business development services or
business environment reforms. This
categorization is a new approach to
the quest for standardized practice. It
avoids the restrictions of standardized
indicators, but still allows for different
projects to be held to account in a
comparable and rigorous way.
Finally, the project has tried to respond
to the fundamental problems of two
approaches to evaluating M4P projects
by combining them both in a new way.
Typical M4P monitoring and evaluation
practice is to focus on multiple
intervention-specific studies which
cannot easily deal with displacement
and deadweight effects – that is,
it cannot easily tell us whether the
economic sector a project is working in
(e.g. ‘Nigerian construction’, ‘Kenyan
tea’) is really improving overall as
a result of the project. Sector wide
studies, with broad baselines using
a randomized sample of firms, on

the other hand, cannot easily deliver
attribution between any sector change,
and an M4P type intervention. This
Handbook, and the GEMS project,
proposes both, combined in such a
way to mitigate the limitations of each
approach.
Several questions are unanswered,
not least the cost trade-off of using
the ‘3 step approach’. More work on
definitions and categorization will
be required to ensure consistency
of usage. We therefore welcome the
engagement of colleagues around
the world as we see if the approach
will have broader benefits for projects
beyond Nigeria. We are already
grateful for the work from ITAD, and
for the insightful comments from
colleagues across DFID, service
provider companies, DCED and the
World Bank. Subsequent iterations
of this Handbook will benefit from an
even broader consultation.
Richard Sandall
Lead Adviser, GEMS
December 2012


Abbreviations

BEE

Business Enabling Environment

BMO

Business Membership Organisation

DCED

Donor Committee on Enterprise Development

DFID

UK Department for International Development

FTE

full time equivalent

GBP

Great British Pound

GEMS

Growth and Employment in States

HNLSS

Harmonised Nigeria Living Standard Surveys

ICT

information and communications technology

ILO

International Labour Organisation

LoCC

League of Construction Companies

M&E

monitoring and evaluation

M4P

Making Markets Work for the Poor

NBS

National Bureau of Statistics

SMEDAN Small and Medium Enterprises Development Agency of Nigeria

iii

1


1

HANDBOOK
PURPOSE

1. Handbook purpose

1


HANDBOOK PURPOSE

The purpose of this handbook is
to provide GEMS with a consistent
framework for capturing and reporting
results and to ensure a coherent
approach to results measurement
across the programme.
DFID is required to report the results of
GEMS to a wide range of stakeholders.
It is therefore important that there
is consistency in the definition and
measurement of its achievements – to
ensure that the results reported are
credible and consistent, and that there
is transparency in the measurement
approaches that are applied. The
application of this handbook is also
intended to provide programme
managers with a consistent framework
to measure success and value for
money and thereby provide information
to guide management decisions
such as whether interventions require
amendment or are worthy of scaling up.
The handbook provides definitions
and measurement standards, taking
into account the wide spectrum of
interventions included under GEMS.
Whilst common definitions and
approaches have been developed, the
handbook has been designed to allow
for flexibility to tailor approaches to
individual interventions and contexts.
The handbook does not provide
comprehensive guidance for every
type of intervention; project managers
are required to adapt the guidance
to their context, whilst ensuring
that they maintain consistency with
the definitions and measurement
approaches provided.

The GEMS results measurement
framework requires assessing changes
that are attributable to an intervention.
It is important that the GEMS deals
with attribution – i.e. drawing causal
links and explanatory conclusions
about the relationship between
observed changes (whether anticipated
or not) and specific interventions.
Attributing if, how, and how much a
given intervention ‘caused’ a particular
‘effect’ are some of the most important
questions for M&E, and some of the
most difficult to answer. The handbook
provides guidance on how to address
these challenges.
The approach builds on, and is
designed to be consistent with,
the DCED Standard for Measuring
Achievements in Private Sector
Development (DCED 2010)1. It should
be used in conjunction with the DCED
Standard guidelines, which contain
more detailed guidance than this
handbook on some aspects of the
results measurement process. However,
the handbook goes beyond the DCED
Standard requirements in several
significant ways:
• The handbook introduces a ‘3-Step
approach’ to results measurement
which combines ‘bottom-up’
and ‘top-down’ measurement
to take account of displacement
and deadweight loss in results
measurement.
• GEMS requires the measurement
of income outreach and value at
the individual or household level,
which goes beyond the minimum
requirements of the DCED
Standard’s ‘Universal Indicators’.
• This framework requires that a
percentage is applied to scale results
according to the estimated level of
attribution that can be assigned to
an intervention. The DCED Standard
simply requires that contributions of
other publicly funded programmes
are acknowledged.

1. http://www.enterprise-development.org/page/measuring-and-reporting-results

The handbook is structured as follows:
Section 2

Strategic framework for
GEMS.

Section 3

Guidance on the
measurement approach
for GEMS components.

Section 4

Standard definitions
of the GEMS outcome
and impact indicators.

Section 5

Selecting measurement
approaches.

Section 6

How to address the
attribution problem.

Section 7

Data sources
and collection
methodologies.

2

3


2

GEMS
STRATEGIC
FRAMEWORK

2. GEMS strategic framework

2


GEMS STRATEGIC FRAMEWORK

GEMS is a partnership between the
Federal Government of Nigeria, DFID
and the World Bank, which aims to
create income opportunities and jobs
for poor Nigerians. GEMS is engaging
in six sectors of the economy2 and has
an additional cross-cutting component
which targets wider policy reform and
strengthening of the business enabling
environment. The seven components of
GEMS are listed in Box 1.
In each component, GEMS is
applying the ‘Making Markets Work
for the Poor’ (M4P) approach.3 The
approach is based on recognition
that economic poverty is the result
of the structure of market systems
in which poor participate. When
markets work efficiently and produce
equitable outcomes for the poor, they
are a powerful vehicle for delivering
growth and poverty reduction. The
approach aims to sustainably improve
the lives of the poor by analysing
and influencing market systems that

affect them as business people (in
terms of higher margins, increased
volumes and improved market access),
consumers (in the form of better access
to products and services, lower prices
and wider choice) and employees (in
the form of higher wages and improved
working conditions). It works to identify
the underlying causes, instead of
symptoms, of why markets do not
work for the poor. Its actions facilitate
change to the behaviour, capabilities,
incentives and relationships of market
actors in order to:
• improve target market systems, and
• create the conditions for markets to
be continuously strengthened after
the M4P ‘intervention’ is completed.

Box 1: GEMS components
GEMS 1 Meat and leather
GEMS 2 Construction and real
estate
GEMS 3 Business enabling
environment and policy
reforms
GEMS 4 Wholesale and retail
trading
GEMS 5 Hospitality and tourism
GEMS 6 Entertainment
GEMS 7 Information and
communications
technology (ITC)

The strategic framework for M4P
interventions is summarised in the
simplified results chain in Figure 1. This
illustrates the links between the logic of
M4P interventions and a typical DFID
logframe4.

Figure 1: M4P strategic framework
M4P strategic framework

DFID logframe level

Changes monitored
• Increased income
• Increased employment
• Outreach

Poverty reduction

Impact

Improved access and growth

Outcome

• More appropriate goods and services
available and taken up by core actors
• Firm performance, including increased
sales

Output

• Improved delivery of support market
services
• Changes in practices of system players
• Crowding in of system players

Input

• Programme undertakes to:
> Alter attitudes of key players
> Encourage linkages between players
> Strengthen capacity and practices

Market system change

Intervention

2. As of December 2012, only GEMS 1–4 are fully operational.
3. For further details about the M4P approach, see www.m4phub.org
4. In addition to the logframe for GEMS components, intervention-specific results chains have been developed which include more levels and capture
additional intervention-specific indicators. In results chains, one can better capture both quantitative and qualitative indicators of each single change. This is
recognized and explored in Section 7 where several ‘generic’ results chains and associated measurement tools are developed for GEMS interventions.

4

5



The GEMS impact indicators are
defined consistently for all components
(see Table 1). These indicators have
some similarity with the DCED Standard
‘Universal Indicators’. However, they
are not fully consistent, and the
definitions used by GEMS imply a
more ambitious level of measurement
than the minimum standard specified
by the DCED (see Table 2). The most
important difference is that GEMS aims
to measure income outreach and value
at the individual or household level,
which goes beyond the DCED Standard
of measurement at the enterprise level.
Detailed definitions of terms used in
defining the indicators are provided in
Section 4.

The application of this framework to
GEMS is presented in Figure 2, which
encapsulates the output to impact
levels of the GEMS logframe5. A
summary of the 10 March 2012 version
of the aggregated GEMS logframe
(encapsulating the benchmarks from
GEMS 1–3) is presented in Annex 1.

The GEMS outcome indicators
are defined in a broadly consistent
manner across the programme (see
Table 3). GEMS 3 applies a slightly
different results framework. The overall
GEMS logframe, and GEMS 1 and 2
logframes, categorize results at four
levels (inputs; outputs; outcome; and
impact).

Figure 2: GEMS strategic framework
Impact
Increase growth, income and employment,
especially for poor men and women, in target markets
in selected states and nationally

Income
(value and outreach)

Employment

Outcome
Improve the performance and inclusiveness of
selected market systems that are important
for poor people

Firm growth

Systemic change
(private sector)

Systemic change
(public sector and
civil society)

Output 1
Inputs, products, services that
benefit the poor at scale … are
successfully established through
market mechanisms (or through
sustainable public funding)
Enterprises
benefitting
from new
inputs,
products and
services

New services
introduced

Improved business
environment

Improved product
quality

Output 2
Key stakeholders pursue more
systemic approaches to economic
development

Improved
understanding
of systemic
approaches to
development

Changed
policies or
programmes

5. Figure 2 encapsulates the cross-GEMS logframe. It does not include the additional ‘intermediate impact’ level that is included in the GEMS 3 logframe.


Table 1: GEMS impact indicators

Impact

To increase growth, income and employment,
in selected states and nationally.

Impact indicator

Indicator definition

1: Income (outreach)

Number of people recording positive change in incomes.

2: Income (value)

Aggregated change in cumulative income (GBP).

3: Employment

Change in employment (FTE jobs).

Source: GEMS logframe

Table 2: Comparison of DCED Standard and GEMS impact indicators
DCED Standard ‘Universal Indicator’

GEMS impact indicator

Scale: Number of target enterprises who
realize a financial benefit as a result of
the programme’s activities per year and
cumulatively. The programme must define
its ‘target enterprises’.

Income (outreach): Number of people
recording positive change in incomes.

Net income: Net additional income
(additional sales minus additional costs)
accrued to target enterprises as a result of
the programme per year and cumulatively.
In addition, the programme must explain
why this income is likely to be sustainable.

Income (value): Aggregated change in
cumulative income (GBP).

Net additional jobs created: Net
additional, full time equivalent jobs created
in target enterprises as a result of the
programme, per year and cumulatively.
‘Additional’ means jobs created minus jobs
lost. ‘Per year’ comprises 240 working days.
The programme must explain why these
jobs are likely to be sustainable. Jobs saved
or sustained may be reported separately.

Employment: Change in employment
(FTE jobs).

Table 3: GEMS outcome indicators

Outcome

To improve the performance and inclusiveness of
selected market systems that are important for poor
people.

Outcome indicator


4: Firm growth (outreach)*

Number of firms (including self employed) with
increased sales

5: Firm growth (value)*

Increase in sales amongst targeted firms

6a: Systemic change and
sustainability: private
sector

Percentage of new or improved products and services,
introduced through project facilitation, that are
established in the market 12 months after project support
has ended

6b: Systemic change
and sustainability: public
sector and civil society

Percentage of new or improved regulations or reforms,
introduced through project facilitation, that are
established in the market 12 months after project support
has ended

7: Product quality

Improvements in product quality (variously defined by
component)

* Defined at the ‘intermediate impact’ level for GEMS 3.


6

7


GEMS 3 (business environment)
has an additional level in its results
framework and the hierarchy flows
as follows: inputs; outputs; outcome;
intermediate impact; impact. Some
of the indicators (numbers 4 and 5)
defined at the outcome level for the
other components are categorised
as ‘intermediate impact’ indicators
for GEMS 3. Other outcome level
indicators (numbers 6a, 6b and 7) are
defined at the same (outcome) level for
all components. GEMS 3 has additional
indicators at the intermediate impact
and outcome levels that are intended
to encapsulate the individual nature
of the GEMS 3 results framework (see
Table 4).


Section 4 provides definitions of all
GEMS indicators defined at the impact,
intermediate impact and outcome
levels: i.e. Indicators 1–7 in Table 1 and
Table 3 and Indicators A–E in Table 4.

Table 4: Additional GEMS 3 indicators
Intermediate impact indicator


A: Increase in firm level investment –
outreach

Number of firms which invest

B: Increase in firm level investment –
rate of increase

Percentage investment made by firms

C: Increase in retained earnings in
targeted firms

Percentage increase over baseline of
retained earnings

Outcome indicator


D: Improved access to land, tax and
investment services

Number of land registration, tax or other
relevant targeted certificates received by
target group

E: Improved Doing Business rating

Percentage improvement in Nigeria’s
absolute Doing Business rating



3

MEASUREMENT
APPROACH

8


MEASUREMENT APPROACH

This section provides guidelines that
have been designed to address the
significant measurement challenges
that are associated with M4P
programmes in order to provide clear
results based on credible evidence.
The effectiveness of GEMS
interventions depends on complex
interactions and feedback loops within
their results chains. These various
feedback loops often make it very
difficult to follow and measure simple
linear causal chains along a onedimensional intervention logic from
intervention to impact.
This suggests a need to split the
measurement process into two
parts to focus the assessments on
relatively contiguous relationships
in the results chain: i) upwards from
GEMS interventions to outputs and
outcomes at the market level; and
ii) downwards from impact-level
development results targeted by
GEMS to outputs at the market level.
This implies the application of a
combination of ‘bottom-up’ and ‘topdown’ approaches to measuring and
attributing change (see Figure 3).
Step 1: bottom-up measurement:
intervention-based monitoring is
applied to assess whether and how
interventions have achieved market
level change and the extent to which
this has led to an improvement in the
performance and inclusiveness of the
market system. It analyses the effects of
interventions and their interaction with
the context, including other market
players and policy inputs. It measures
the degree to which intervention
activities have achieved their intended
(and unintended) outputs and attempts
to measure the degree to which
intended outcomes have resulted.

Figure 3: A 3-Step approach to measuring change

Impact
(poverty level)

Impact
(sector level)

Impact
(enterprise level)

Outcome
(enterprise/core
market)

STEP 3

3.1. A 3-Step approach
to measuring and
attributing change

STEP 2

3

3. Measurement approach

STEP 1

9

Output
(service/support
market)

Intervention

Step 2: top-down measurement:
this step assesses the key changes in
indicators at the impact level; and then
assesses the factors at the market level
that have driven this change.
Step 3: compare and triangulate
findings: this step analyses logical
relations between GEMS interventions
and development results (i.e.
impact level indicators). This entails
synthesizing and cross-checking the
results of the previous two steps by
focusing on the levels where the two
steps come together at the output and
outcome levels. It assesses the extent

to which the outputs and outcomes
achieved by GEMS interventions (which
are measured in Step 1. are consistent
with the market-level factors that
have driven changes in impact level
indicators (measured in Step 2. This
step allows summarising, synthesizing
and double-checking of results. It is
likely to be carried out through an
iterative process by:
i) synthesizing the results of Step 1 and
Step 2, with a view to ensuring their
comparability; and
ii) comparing the actual ‘transitive
relationship’ between GEMS
interventions and the impact-level
results that they seek to achieve.


Step 1 broadly represents the typical
M4P intervention-specific measurement
approach. Step 2 is the equivalent of
a sector wide survey, using a random
sample of firms within a sector to
detect sector wide changes that
an intervention specific evaluation
cannot provide (at least without
risking overlooking deadweight and
displacement). Step 3 is the process of
further study that brings the results of
steps 1 and 2 together.
Because this process includes
both bottom-up and top-down
measurement of change between
the output and outcome levels, it
provides triangulation of the evidence
of change at this level. It also provides
a reasonably robust approach to
measuring attribution and minimizing
self-importance bias.6 At the same time
it recognises the difficulty of assessing
the influence of GEMS interventions
at outcome and impact levels without
a broader understanding of sectoral
social and economic performance.
The 3-Step approach is most likely
to provide rigorous evidence that
addresses biases inherent in the
application of either the Step 1 or
Step 2 approach on their own (see
Box 2). The approach should enable
displacement and deadweight loss
associated with an intervention to be
taken into account: Step 2 involves
measurement of overall change to
impact and outcome indicators which
can be used to contextualise changes
observed as a result of interventions
through Step 1.
In the actual application of this
approach, the scope and depth of the
Step 2 – and its relative importance
vis-à-vis Step 1 – will depend on the


Box 2: Measurement challenges
Bottom-up measurement

Top-down measurement

Risk of double counting impact
across interventions.

Very challenging/costly to make
surveys and quantitative analysis
representative.

Risk that measurement will ignore
deadweight loss and displacement.

Large attribution challenges – big
jump between micro interventions
and macro economy-wide impacts.

Difficult to account for impact
of synergies across different
components of a programme.
Risk of ‘self-importance bias’ in
estimating attribution.
Large attribution challenges – big
jump between micro interventions
and macro economy-wide impacts.

specific nature of interventions and the
extent to which external factors need
to be analysed in measuring the impact
of GEMS. It may be possible in some
cases to derive preliminary estimates
of impact through Step 1, although
the subsequent steps will normally
be required to ensure that estimates
take account of displacement and
deadweight loss, and are triangulated.

the measurement process. It aims to
answer the following four questions:

Splitting the results measurement
process into two separate steps
(plus a third for crosschecking and
conclusions), allows introduction of
a significant methodological and
practical distinction between the
steps and permits the use of more
appropriate methods for each part of

• How significant were each of these
chains of influence?

• Have the key outcomes and impacts
targeted by GEMS improved over
the measurement period?
• What were the determining factors
of those changes?
• Were those determining factors
in turn influenced by GEMS
interventions?

The appropriate measurement tools
for addressing each question will vary
depending on the intervention and
the context. Examples are provided in
Annex 5.

6. This is arguably a common bias when measuring attribution in international development programmes: the relative importance of other contextual factors is
often under-estimated. White and Phillips (2012) describe ‘self-importance bias’ as follows: “People are indeed the centre of their own universe; the problem
comes when they think that they are also the centre of everyone else’s universe, and as a result they overstate their role in events...it has been found that it is
domestic political processes that actually drive policy change… so any account which ascribes a central role to external actors is likely to be overstating the
importance of those actors.”

10

11


3.2. Six stages of
measurement
The guidance provided in this
handbook focuses on the higherlevel effects from GEMS interventions
(at the impact and outcome levels).
Approaches to measuring these higherlevel effects should be careful to take
account of the following commonly
agreed requirements in impact
assessment:
• Measurement should estimate a
given effect (on final beneficiaries)
from a particular cause.
• It is important to measure attribution,
suggesting a direct link between a
cause and an effect.
• Measurement should take account
of the counterfactual – i.e. asking
what would have happened without
the intervention in making causal
inference.
In addressing these factors, the GEMS
results measurement approach follows
six stages listed below, which are further
elaborated in the remainder of this
section. Each stage should apply the
3-Step approach. For example, the
‘research questions’ should be tailored
to address bottom-up and top-down
measurement questions; measurement
methods differ for each step (Stage 5);
and Step 3 requires specific indicators
for Steps 1 and 2.
Stage 1: Articulate the results chain
For each GEMS component, and
for specific interventions within the
component, managers should establish
and, wherever possible, agree with
stakeholders a ‘plausible’ results chain
that accurately reflects the ways in
which GEMS will deliver or enhance the
delivery of planned results. Discussions
should focus on the type and nature of
cause and effect relationships at each
stage in the results chain, including:
• Direct control: where the GEMS
intervention has fairly direct control
of the results, typically at the output
level.

Stage 1: Articulate
the results chain

Stage 2: Define the
‘research questions’


• Direct influence: where the GEMS
intervention has a direct influence
on the expected results, such as the
reactions and behaviours of its target
groups through direct contact.
• Indirect influence: where the GEMS
intervention can exert significantly
less influence on the expected results
due to its lack of direct contact with
those involved and/or the significant
influence of other factors. Channels
of indirect influence include:
> Crowding in: other service
providers start applying
the practices of impacted
programme ‘beneficiaries,’ by
seeing the positive impact of
programme activities on them.
> Sector growth: as a result
of programme activities, the
sectors in which it works grow;
existing enterprises expand their
businesses while ‘new entrants’
come into the market.
> Backward and forward linkages:
changes at one point of the
market trigger changes at other
points along the value chain.
Because of the means by which
GEMS seeks to achieve change, its
interventions will frequently produce
significant indirect and long term, and
less easy to measure results.
Stage 2: Define the ‘research
questions’
The key questions and hypotheses to be
tested during the results measurement
process (the ‘research questions’)
should be generated from the causal
models articulated in the results chains,
covering both Step 1 and Step 2 of the
measurement process. Figure 4 shows
where the 3 Steps can be applied in
the generic GEMS strategic framework.
The ‘generic’ research questions for
GEMS can be drawn from this (see Box
3). More specific research questions
will need to be tailored to individual
interventions.

Stage 3: Define
indicators of change

Stage 4: Establish
measurement methods

Stage 3: Define indicators of change
The impact and outcome indicators
of change, whose measurement this
handbook focuses on, are already
defined in the GEMS logframe.
Indicators associated with the
intervention-specific results chains
should, wherever appropriate, be
consistent with these indicators,
although it is likely to be the case
that additional intervention-specific
indicators will need to be developed.

Box 3: ‘Generic’ research
questions for GEMS
Step 1
• What have been the
direct outputs from GEMS
interventions?
• What changes have occurred
at the service/support market
level? To what extent can these
be attributed to GEMS? What
have been the other causal
factors?
• To what extent have the service/
support level market changes
led to an improvement in the
performance and inclusiveness
of the core market system?
Step 2
• How have the level and
composition of income and
employment changed (in the
relevant sector/state)?
• What market-level factors have
led to the changes in income
and employment?
• What have been the causes of
any changes in market level
factors?
Step 3
• To what extent can the market
level changes observed in
Step 2 be attributed to GEMS
interventions?

Stage 5: Measure
changes in indicators

Stage 6: Estimate
attributable change



Figure 4: Applying the 3-Step approach to the GEMS logframe

Impact

Increase growth, income and employment,
in selected states and nationally

Income

Employment

Outcome

Systemic
change
(public
sector
and civil
society)

Output 1

Inputs, products, services that
benefit the poor at scale … are
successfully established through
market mechanisms (or through
sustainable public funding)
Enterprises
benefitting
from new
inputs,
products and
services

New services
introduced

Improved
business
environment

Improved
product
quality

STEP 3

Systemic
change
(private
sector)

STEP 2

Firm
growth
(value and
outreach)

STEP 1

Improve the performance and inclusiveness of
selected market systems that are important
for poor people

Output 2

Key stakeholders pursue more
systemic approaches to economic
development

Improved
understanding
of systemic
approaches to
development

Changed
policies or
programmes

Interventions

Stage 4: Establish measurement
methods

Stage 5: Measure changes in
indicators

Measurement methods should be
based on an evaluability assessment.
Different combinations of measurement
approaches will be appropriate for the
three steps. Further guidance on the
selection of measurement approaches is
provided in Section 5.

Measurement should allow for analysis
of ‘impact heterogeneity’, including
sub-group analysis by gender and
poverty status as appropriate for the
intervention. It should capture both
positive and negative effects. Annex 5
provides guidance on how changes in
indicators over time can be measured
for specific types of intervention.
Data sources and collection methods
are considered in Section 7.

7. http://www.enterprise-development.org/download.aspx?id=2012

Stage 6: Estimate attributable change
Stage 6 involves an examination of
the validity of the causal linkages in
the results chain and the evidence to
support them, which should be used
to build a story of plausible attribution.
Based on this analysis, a percentage
should be assigned to the extent to
which the measured change can be
attributed to the GEMS intervention.
Further guidance is provided in Section
6 and by the DCED.7

12

13


4

DEFINING
INDICATORS
OF CHANGE

4. Defining indicators of change

4


DEFINING INDICATORS OF CHANGE

This section provides guidance on
the definition and measurement of all
GEMS impact and outcome indicators.
In addition, the table below4 provides
guidance on the definition and
measurement of poverty – because the
GEMS results framework requires that
impact indicators are disaggregated
between poor and non-poor.

Indicator

Poverty

Definition

The poverty line is defined as ₦66,000 per capita in 2010 prices8.
In 2012 prices, this is equivalent to ₦82,4049. For each region, the
General Household Survey data presented in Annex 4 should be
applied to derive from this an estimate of the average level of
income per worker required to keep his/her household above the
poverty line.
For example, for Kano (which is in the NW Zone), the average
household size is 6.75, meaning that the income required to keep the
average household above the poverty line in 2012 is:
₦82,404 x 6.75 = ₦556,227
45.3% of the household is of a working age of 15 to 65, meaning that
on average, each worker will require a net annual income10 of:
₦556,227 / (6.75 x 0.453) = ₦181,907
Equivalent calculations for other zones in Nigeria are provided in
Annex 4.

Units of
measurement

The unit of measurement is the annual net income per working
person. To derive this figure from a daily wage, it should be assumed
that a working person engages in productive work for 240 days per
year.
Income is defined as the amount of money received for labour, for
services, from the sale of goods or property, or from investments.
Non-cash income should be assigned a monetary value at current
market prices and included in income calculations.

Exclusions

n/a

8. This is based on World Bank (2010), which applies several methods to derive the absolute poverty line for Nigeria using data from the HNLSS. The results
derived were consistent: at around ₦66,000 per annum in 2010 prices. This is calculated by combining the cost of a national food basket required to deliver
3,000 Kcal per person to day and adding a non food component.
9. Applying a deflator derived from the average All Items CPI for January to August 2010 and January to August 2012.
10. i.e. annual income after tax.

14

15



Indicator 1

Income (outreach)

Definition

Net number of income earners recording a positive change in
incomes.
Where an individual receives income from multiple sources, the unit
of measurement is the earnings from the income stream targeted by
the GEMS intervention.
Non-cash income should be valued at current market prices and
included in income calculations.
A positive change in income is defined as 15% or more in real Naira
terms within the GEMS component implementation period.
An increase in income should be additional, which is defined by
the HMT Green Book as ‘an impact arising from an intervention
is additional if it would not have occurred in the absence of the
intervention’.

Units of
measurement

Disaggregate between:
• Formal and informal. See Annex 2 for a definition of informal
employment.
• Employees and self employed/family business members.
• Male and female.
• Poor and non-poor.
• Direct and indirect impacts. See page 11 for further specification
of the distinction.
Take account of deadweight loss in calculations. Displacement
should be considered11 and, wherever possible, calculated (through
the 3-Step approach). As a minimum, an evidence-based statement
of possible displacement effects should be included as part of
the results measurement process. The geographical limit for
displacement estimates is Nigeria.

Exclusions

n/a

11. The DCED Standard requires that programmes should “cite or produce evidence that displacement has been taken into account in the development of the
results chain(s)”. “Research should consider likely displacement within and outside the value chain i.e. where non target groups suffer because the target group
benefits. To assess this, programmes will need to consider whether the markets are involved are shrinking, static or growing. Displacement will be far higher in
shrinking and/or saturated markets”.


Indicator 2

Income (value)

Definition

Net aggregated change in cumulative income (Naira) accrued by
individuals as workers.
Non-cash income should be valued at current market prices and
included in income calculations.
Includes both people who enter the sector through job creation and
people who are already in the sector.
An increase in income should be additional, which is defined by
the HMT Green Book as ‘an impact arising from an intervention
is additional if it would not have occurred in the absence of the
intervention’.

Units of
measurement

Average over a twelve month period.
• Formal and informal. See Annex 2 for a definition of informal
employment.
• Poor and non-poor.
• Direct and indirect impacts. See page 11 for further specification
of the distinction.
should be considered and an evidence-based statement of possible
displacement effects should be included as part of the results
measurement process. The geographical limit for displacement
estimates is Nigeria.

Exclusions

Exclude unpaid family labour.
Apprenticeships should be excluded from the calculations except for
where there is evidence that the apprenticeship has been established
with the primary purpose of addressing a labour shortage rather than
to fulfil a training need. Where included, apprenticeship calculations
should be disaggregated from other income generating activities.


16

17



Indicator 3

Employment

Definition

Change in employment (FTE jobs)
Net additional, full time equivalent jobs created in target enterprises
as a result of the programme, per year and cumulatively by persons
over the age of 15.
‘Additional’ means jobs created minus jobs lost.
‘Per year’ comprises 240 working days of eight hours.
A job is sustained when the position, or FTE equivalent, has been
in existence for 12 months or more. The programme should explain
why jobs are likely to be sustainable beyond the lifetime of the
intervention.
Jobs saved or sustained may be reported separately.
Only include work that is paid at a minimum in excess of the poverty
line (see poverty indicator above). Non-cash income should be
valued at current market prices and included in income calculations.

Units of
measurement

Full time equivalent (FTE). Seasonal jobs may be counted so long as
they are included on a pro rata basis (i.e. a 3 month full time job =
0.25 FTE).
The maximum FTE value for a single job is 1, even if more than 240
days (or 8 hours per day) are worked.
A job should not be counted again if it is filled over time by different
people.
Disaggregate by:
• Poor and non-poor: estimate proportion of jobs that are created
which are provided to people who were ‘poor’ before entering
employment.
• Formal and informal employment: see Annex 2 for a definition of
informal employment.
• Direct and indirect impacts. See page 11 for further specification of
the distinction.
should be considered and an evidence-based statement of possible
displacement effects should be included as part of the results
measurement process. The geographical limit for displacement
estimates is Nigeria.

Exclusions

Jobs are generally excluded where associated with businesses
moving from other parts of Nigeria. An exception to this is where
the relocation can be demonstrated as genuinely safeguarding the
job(s) concerned (i.e. the job(s) would have been lost at the original
location).
Exclude unpaid family labour, forced labour and
child labour.
Apprenticeships should be excluded from the calculations except for
where there is evidence that the apprenticeship has been established
with the primary purpose of addressing a labour shortage rather than
to fulfil a training need. Where included, apprenticeship calculations
should be disaggregated from other income generating activities.


Indicator 4

Firm growth (outreach)

Definition

Net number of firms (including-self employed) with increased sales.
Sustained increase in sales is when a business can demonstrate an
increase to its turnover over a minimum of a 12 month period.

Units of
measurement


• Micro, small, medium and large enterprises (following SMEDAN
definitions12 ).
• Formal and informal.
• Male/female managed or owned firms.
• Direct and indirect effects.

Exclusions

n/a

Indicator 5

Firm growth (value)

Definition

Net annualised increase in sales amongst firms (Naira).

Units of
measurement

definitions13 ).
• Direct and indirect effects. See page 11 for further specification of
the distinction.

Exclusions

n/a

Indicator 6a

Systemic change and sustainability: private sector

Definition

Percentage of jobs and income opportunities created by GEMS that
can be attributed to private sector interventions that are sustainable.

Units of
measurement

‘Intervention’ is defined as “an activity or series of activities designed
to achieve a specific change in the support functions/market or a set
of activities designed to achieve the sustainable delivery of a new
or improved service or output that, through its use by the target
group, will result in improved business environment and contribute
to increased incomes for that target group”.
‘Private sector intervention’ is defined as “improved products,
services, relationships and technologies introduced through project
facilitation”.
‘Sustainable intervention’ is defined as one that remains established
in the market 12 months after project support
has ended.

Exclusions

n/a

12. & 13. SMEDAN applies the following definitions: micro enterprise – up to 9 employees; small enterprise – 10–49 employees; medium enterprise – 50–199
employees; large enterprise – over 200 employees.

18

19



Indicator 6b

Systemic change and sustainability: public sector and civil
society

Definition

Percentage of jobs and income opportunities created by GEMS that
can be attributed to public sector and civil society interventions that
are sustainable.

Units of
measurement

‘Intervention’ is defined as “an activity or series of activities designed
to achieve a specific change in the support functions/market or a set
of activities designed to achieve the sustainable delivery of a new
or improved service or output that, through its use by the target
group, will result in improved business environment and contribute
to increased incomes for that target group”.
‘Public sector and civil society intervention’ is defined as new or
improved regulations or reforms introduced and established that are
enabled or facilitated through GEMS, and improve market conditions
and promote inclusiveness.
‘Sustained intervention’ is defined as one that remains established in
the market 12 months after project support
has ended.

Exclusions

n/a

Indicator 7

Improvements in product quality

Definition

Improvements in product quality defined individually by components.

Units of
measurement

n/a

Exclusions

n/a

GEMS 3-specific indicators
Indicator A

Increase in firm level investment – outreach

Definition

Number of firms which undertake new investment as a result of
improvements in the business environment facilitated by GEMS.
Investment is defined as increase in non-labour assets adjusted for
depreciation and GDP growth.

Units of
measurement

definitions14).
• Direct and indirect effects. See page 11 for further specification of
the distinction.

Exclusions

14. See footnote 11.

n/a



Indicator B

Increase in firm level investment – rate of increase

Definition

Annualised investment made by firms as a result of improvements in the
business environment facilitated by GEMS.
Investment is defined as increase in non-labour assets adjusted for depreciation
and GDP growth.

Units of measurement

Investment as percentage of turnover.
• Micro, small, medium and large enterprises (following SMEDAN definitions).
• Direct and indirect effects. See page 11 for further specification of the
distinction.

Exclusions

n/a

Indicator C

Increase in retained earnings in targeted firms

Definition

Net annualised increase in retained earnings as a result of improvements in the business
environment facilitated by GEMS.
‘Retained earnings’ is defined as net earnings not paid out as dividends, but retained by the
company to be reinvested in its core business or to pay debt.
Retained earnings is calculated by adding net income to (or subtracting any net losses from)
beginning retained earnings and subtracting any dividends paid to shareholders:
Percentage increase over baseline of retained earnings.

Units of
measurement

• Micro, small, medium and large enterprises (following SMEDAN definitions).
• Direct and indirect effects. See Paragraph 25 for further specification of the distinction.

Exclusions

n/a

Indicator D

Improved access to land, tax and investment services

Definition

Number of land registration, tax or other relevant targeted certificates received
by target group.
To be further defined by GEMS 3 as interventions evolve.


Compliance certificates, land registration certificates and certificates of
registration as a formal company.

Exclusions

n/a

Indicator E

Improved Doing Business rating

Definition

Percentage improvement in Nigeria’s absolute Doing Business rating
Notes: indexed by GEMS 3; baseline = 100


See http://www.doingbusiness.org/methodology
Disaggregate by each of the ten doing business categories used in the rating.

Exclusions

n/a

20

21


5

SELECTING
MEASUREMENT
APPROACHES

5. Selecting measurement approaches

5


SELECTING MEASUREMENT APPROACHES

5.1. Applying a mixed
methods approach

• The extended and complex
nature of the impact pathways
in GEMS. As acknowledged in the
DCED Standard, demonstrating that
interventions cause development
effects depends on providing
evidence to verify the causal claims
made in the results chain of the
programme or intervention.

As a result, in most cases, a theorybased measurement approach that
applies mixed methods is likely to be
the most suitable – both in providing
convincing evidence of impact and in
explaining why changes occurred and
how they varied across circumstances.
It is important to try to spell out a
counterfactual and rule out alternative
explanations as much as possible.
Where a counterfactual scenario cannot
be made, credible impact assessment
needs to be built around the results
chains of the programme and individual
interventions – that allows users to
demonstrate the causal links proposed.

• It is often not possible to fully
isolate the effect of a GEMS
intervention from the possible
multitude of factors which might
have an influence on the outcome of
interest. Most GEMS interventions
are ‘contributory causes’: they
work as part of a causal package
in combination with other ‘helping
factors’, including stakeholder
behaviour, related programmes and
policies, institutional capacities,
cultural factors, and socio-economic
trends.

Rather than selecting one method,
programmes should aim to use a range
of tools to collect and analyse the
necessary data; information generated
by mixed methods can help to establish
the validity of the data and the
reliability of the measures of change.
However, it is not necessary to use a
different tool for each indicator, in fact,
it is important to group the indicators
together and collect data on as many
as possible with the same tool or tools.
This will make the data collection both

In measuring the results of GEMS
interventions, two common factors
need to be taken into account:

manageable and efficient. Programmes
should balance the use of appropriate
tools with keeping the overall results
measurement process affordable and
manageable.
The measurement of different
stages in the results chain will be
amenable to different measurement
approaches. A mixture of quantitative
and qualitative data is likely to be
required to build credible evidence
of the links and attribution between
GEMS interventions and impact-level
indicators of employment and income
generation. The tools applied should
include some combination of the
methods listed in Box 4.
The sheer complexity of GEMS
interventions means that there will
always be limits to the accuracy of
estimates of impact. This does not
suggest that results measurement is not
valuable. But it does imply that caution
is required in expectations of precision.
It is important that GEMS is transparent
about the margins of error in its results
measurement.

Box 4: Menu of measurement methods
1. Quantitative methods
• Before versus after the intervention.
• ‘With and without’ comparison.
• Quasi-experimental approaches such as difference in difference analysis.
• Regression analysis.
• Experimental methods such as randomized controlled trials.
2. Qualitative methods
• Key informant interviews (KII) to examine the quality of the results achieved, the main
contributing factors to this achievement, the validity of the results chain and the relationship
between risks, assumptions and performance.
• Focus groups with beneficiaries and specific target groups (as for KII).
• ‘Comparative case studies’ to assess ‘with and without’ intervention scenarios.
• ‘Outcome Harvesting’ to collect evidence on what has been achieved and work backwards
to determine whether and how the project or intervention contributed to this change.
• ‘Most Significant Change’ to collect stories emanating from the field level, and then
the systematic selection of the most significant of these stories by panels of designated
stakeholders allowing whole teams of people to focus their attention on programme impact.

22

23


5.2. Selecting
measurement methods
The overall results measurement objectives
and the specific research questions that
need to be answered are important
factors in deciding which measurement
approach(es) to use. Measurement
objectives and research questions should
be developed from the results chain. The
appropriate combination of methodologies
will also depend on the nature of the
intervention and the context in which it is
taking place. Considerations include those
set out in Table 5.


Table 5: Evaluability considerations and questions
Considerations

Questions

Objectives

• Does the intervention have clear objectives/
outcomes it aims to achieve?
• Are these aims – or at least some key ones – defined
and measurable?

Counterfactual

• Do we know what is the target population/s is/are? Is
the target population pre-selected or self-selecting?
• Can we identify the population being treated? Are
the characteristics of the treatment group known?
• Can a comparison group with matching
characteristics be created?
• Are members of the comparison group likely to
become members of the treatment group during the
course of the evaluation?

Table 6 provides guidance on how these
evaluability questions should be considered
in the context of the generic research
questions for GEMS. Further guidance
on results measurement for specific
‘intervention types’ is provided in Annex 5.

• Can we identify untreated businesses that may be
affected by spill overs from the treatment?
• Where a comparison group cannot be identified
(which makes ‘with and without’ comparison
unfeasible), what alternative measures of the
counterfactual are feasible?15
Treatment

• Can we specifically define the treatment?
• Are there identifiable mechanisms through which the
results are expected to be attained?
• Is the treatment uniform in relevant respects for all
participants?
• Is the provision of the intervention limited in some
way – by geography, or type of firm?
• What is the number of units in the treatment and
comparison groups?

Complexity

• To what extent are there other ‘contributory causes’
to the achievement of the results?

Significance

• How significant is the intervention in terms of the
identifiable change it represents?

Cost

• Cost and effort needs to be proportionate to the
size of the intervention, its importance, and the value
of the information that the measurement process
will provide.

15. Comparison groups are particularly difficult to identify for complex interventions or when the number of observations or case studies is small (common
circumstances in GEMS). Where this is the case, there are other useful approaches to measuring the counterfactual which, depending on the situation, may be
appropriate: e.g. simulation modeling; cross-sectional comparison; cohort studies; case control studies; looking for frequency of association between cause and
effect; and/or association and analysis of multiple combinations of causes (for example, qualitative comparative analysis).



Table 6: Measurement design and methods selection framework
Step 1 – Intervention to market system change (support and core markets)

EQ1.1

What have been the
direct outputs from
GEMS interventions?

Contextual factors to consider in
measurement

Implications for measurement design and
methods

• ‘Outputs’ in the GEMS logframe and DCED

• Monitoring data, based on intervention

• Measurement should start with the ‘direct’

Research question

• Construction of intervention-relevant baseline

Standard are set at a high level – i.e. at
market-level change that often lies beyond
the direct influence of GEMS.
outputs.

manager observations and key informant
interviews is likely to provide the required
information.
data capturing the initial situation in relation to
beneficiaries prior to the project start.

• Where possible, specification of a

counterfactual – comparison group to facilitate
comparison of ‘with’ project and ‘without’
project scenarios.

• Qualitative data gathering through interviews,

focus group discussions and consultations with
key market players to provide a lens to explore
how the interventions are delivering change.

What changes have
occurred at the
service/support
market level?

• To what extent has there been a change in

To what extent can
these changes in
the service/support
market be attributed
to GEMS?

EQ1.2

• Can we define the treatment and the

the service/support market?

• Market analysis of the service support market
(demand and supply analysis).

• Qualitative data gathering through interviews,

focus group discussions and consultations with
key market players to examine the changes
taking place.

pathways through which the results are
expected to be attained? Challenges may
result from the indirect, dynamic and flexible
nature of many GEMS interventions.

• Can a comparison group with matching
characteristics be created?

• Is there likely to be spill-over and

contamination between the treatment and
comparison group?

• What is the number of units in the treatment
and comparison groups?

• A theory-based measurement approach

requires definition of the treatment and results
pathways. However experimental and quasiexperimental methods can measure change
in the absence of a definition of the results
pathways.

• If feasible, propensity score matching may

enable the construction of a comparison group
required for ‘difference in difference’ analysis.

• Contamination will limit or mitigate

the feasibility of experimental or quasiexperimental methods.

• This will affect the extent to which survey data
is representative.

What have been the
other causal factors?

• To what extent are there other ‘contributory
causes’ to the achievement of the results?

• To what extent are there likely to be multiple
benefits or ‘unintended consequences’?

• How significant is the intervention, in terms
of the identifiable change it represents?

• Other contributory causes need to be

recognised in the results chain and ‘controlled
for’ in impact measurement.

• Measurement design needs to look for and

take account of unintended consequences of
market interventions.

• The significance of anticipated change as a

result of the intervention affects the extent to
which sample sizes are representative. Survey
design needs to take account of this.

EQ1.3

To what extent
have the service/
support level market
changes led to an
improvement in the
performance and
inclusiveness of the
core market system?

• Are the core market systems aims – or

• Aims should wherever possible be defined and

• Are the core market system beneficiaries

• If beneficiaries can be identified and pre-

at least some key ones – defined and
measurable?

identifiable, or is the change more systemwide? Are the beneficiaries pre-selected or
self-selecting?

• Can we define the treatment and the

pathways through which the results are
expected to be attained?

• To what extent are there other ‘contributory
causes’ to the achievement of the results?

• How significant is the service/support level
market change, in terms of the identifiable
change it represents in generating change
in the core market’s performance and
inclusiveness?

measurable as a benchmark against which to
measure change.

selected, experimental or quasi-experimental
methods are more likely to be feasible.

• Challenges may result from the indirect,

dynamic and flexible nature of many GEMS
interventions.

• Other contributory causes need to be

recognised in the results chain and ‘controlled
for’ in impact measurement.

• The significance of anticipated core market

system change affects the extent to which
sample sizes are representative. Survey design
needs to take account of this.

24

25



Table 6: Measurement design and methods selection framework
Step 2 – Changes in income and employment in core market
measurement

methods

How have the level
and composition
of income and
employment
changed (in the
relevant sector/
geographic area)?

• Do results include data on other

• General Household Survey and HNLSS

What have been
the causes of any
changes in market
conditions?

• What are the key drivers of improvements in

Research question
EQ2.1

EQ2.3

characteristics of the population – to enable
matching etc?

• How regular and reliable are secondary data
sources (e.g. household surveys)?

firm performance and employment?

• What are the key trends in the core market
and what are the underlying causes?

undertaken by NBS provide infrequent
estimates of income and employment.
Alternative data sources are likely to be
required.

• Cross-GEMS enterprise survey.
• Qualitative analysis of data on the composition
of economic growth and the evolution of
poverty and inequality.

• Documentary analysis.
• Interviews and focus group discussions.
• Regression analysis/Difference-in-Difference
analysis.

• Employment surveys.
Step 3 – Compare and triangulate findings – linking core market changes to GEMS
Research question
EQ3.1

To what extent can the
market level changes
observed in Step 2 be
attributed to GEMS
interventions?

measurement

methods

• Taking the impact of GEMS on the support

• Triangulation of the findings from Steps 1 and

market, and knowing the changes taking
place in the core market, can we distil with
some degree of confidence what elements of
the changes in the latter are due to GEMS.

• For some GEMS interventions, it may be

possible to use overall intervention to impact
measurement techniques in some cases (i.e.
not the 3-Step process).

• In most cases the 3-Step process will be a
preferable approach due to limitations of
any single method in measuring the links
between interventions and impact.

• The 3-Step process should be used to

carefully assess the extent to which estimates
derived through Step 1 are picking up
displacement and deadweight loss.

2 to build the evidence that the changes in the
services/support market due to GEMS have:
> plausibly resulted in changes in business
performance in the core market which in
turn have
> plausibly resulted in quantifiable changes in
levels of employment and income for poor
people.

• Analysis of the actual ‘transitive’ process

between GEMS interventions and the impactlevel results.

• Application of qualitative methods, such as
Most Significant Change techniques.


5.3. Aggregating results
In aggregating results across
interventions, the following factors
should be taken into account which,
if not dealt with carefully, may
compromise the integrity of the results
reported:
• Many indicators may be defined in
different ways in different contexts.
This may result in the aggregation of
inconsistent units.
• There is a risk of double counting
between interventions and
components – e.g. at the impact
level between employment and
income effects
• By aggregating results from
individual interventions or
components, there is a risk that
the impact of synergies between
programme parts are missed.


Therefore, when aggregating
data, GEMS teams should identify
interventions that have overlapping
beneficiaries and properly account for
this. Component-wide results chains
should be developed to help identify
overlaps between interventions, and
to illustrate the influence of external
causal factors (including other GEMS
components). GEMS should collectively
map the beneficiaries of interventions
and their geographical locations to
identify overlaps. After identifying
overlaps, figures should be corrected
by counting only once the beneficiaries
that have been reached by more than
one intervention.

The DCED offers the guidance in Box 5
on the treatment of overlaps.

The consistent application of the
definitions provided in this handbook,
combined with the 3-Step approach
to measurement should enhance the
integrity of aggregated results.

Box 5: Measurement guidance for overlapping interventions
Adjustment required
Outreach
Overlap less than 5%

Add all beneficiaries (no corrections)

Overlap more than 95%

Account for only the largest number (so
no ‘adding’ at all)

Overlap between 5 and 95%

Estimate each overlap(s) and show
calculation

Income/Jobs
If attributable (isolated) impact per
(cluster of) interventions

Add all beneficiaries

Pilot and upscale phase
Upscale (phase 2) interventions
probably overlap with pilot (phase 1)
interventions if target beneficiaries are
the same
Outreach

Deduct 100% after completion of
upscale intervention (achieved/
projected)

Income/Jobs

‘Freeze’ impact of pilot intervention at
the start of the ‘upscale intervention’

Similarly, due to the way that impact
indicators are defined in GEMS, there
is a strong possibility that the creation
of a new employment opportunity
could be counted twice – as a job
and as an increase in income for an
individual. This would happen, for
example, where an individual who
was previously working informally and
receiving earnings below the poverty
line acquires a newly created job that
moves her above the poverty line.
GEMS components should be aware of
the likelihood of such double counting
and estimate what proportion of jobs
and income outreach estimates that are
reported contain overlaps.

26

27


6

ESTIMATING
ATTRIBUTABLE
CHANGE

6. Estimating attributable change

6


ESTIMATING ATTRIBUTABLE CHANGE16

Attribution involves a causal claim
about the intervention as the cause of
the result, and measurement of how
much of the result can be linked to
the intervention. In the case of GEMS,
interventions operate as part of a wider
system where, in nearly all cases, they
interact in some way with other publicly
funded and private activities to achieve
their intended results.

Figure 5: Attribution – causality and external influence

High

Low
External influence

Most GEMS interventions can therefore
be seen as a ‘contributory’ cause –
i.e. the intervention is a vital part of
a ‘package’ of causal factors that are
together sufficient to produce the
intended effect. However on its, own,
the intervention is neither sufficient, nor
always necessary to achieve a desired
result. This focuses attention to the role
of the intervention in a package
of causes.
GEMS interventions aim to catalyse
change, inducing spill-over effects
to indirectly scale up change. This
means that external factors have an
important influence on the scale and
nature of change, which grows as we
move further up the results chain (see
Figure 5).

Poverty
reduction

Improved
growth and
access

Market system
change

Influence of
intervention
(causality)
Low

Intervention

High

Given the multiple factors that can
affect outcomes and impacts, results
measurement needs to measure
the added value of the intervention
under consideration, separate from
those other factors. Any observed
changes will be, in general, only partly
caused by the intervention of interest.
Other factors will often interact and
strengthen/reduce the effects of the
intervention of interest. Therefore,
addressing this ‘attribution problem’
implies both isolating and accurately
measuring the particular contribution
of an intervention and ensuring that
causality runs from the intervention to
the result.

16. The DCED provides useful guidance on addressing attribution. See http://www.enterprise-development.org/download.aspx?id=2012

28

29


Whilst rigorous proof of attribution
will be beyond the means of almost all
programmes, attribution should always
be demonstrated to a level that would
convince a reasonable but sceptical
observer. GEMS should report on
the other contributors to the change,
as accurately as possible, assign a
percentage to the level of attribution
that can be claimed17. This should be
estimated for each level in the results
chain and reflected in the figures
reported. The basis for the calculation
of this percentage should be presented
transparently.
This should be done according to the
following stages of theory-based results
measurement for each level in the
results chain:18
1. Set out the attribution problem
to be addressed: Assess the
nature and extent of the attribution
challenge by asking:
• What do we know about
the nature and extent of the
contribution expected?
• What other public programmes
and private actions will have
contributed to the changes
claimed?19
• What would show that GEMS
has made an important
contribution?20
• What would indicate that GEMS
has had the effects envisaged
in the results chain for the
intervention?

6. Estimating attributable change

2. Assemble and assess the
contribution narrative and
challenges to it: From the outset, it
is important to validate whether the
results chain and the assumptions
that it depends on hold true.
This validation process should be
undertaken systematically and
regularly in order to iteratively
build up a convincing and plausible
evidence-based narrative on the
effects that GEMS is having in direct
and/or indirect ways. It is desirable
that this process involves relevant
external stakeholders who are in
a position to externally verify that
the original results chain and future
observed changes are plausible
and credible.
3. Gather evidence to verify the
contribution narrative: The type
of evidence gathered will largely
depend on the nature of the
intervention and the context. Further
guidance on specific methods is
provided in Section 5. Ideally the
evidence base will consist of a
combination of quantitative and
qualitative data focused on testing
and proving the results chain.
Qualitative impact studies will
complement quantitative studies,
while case-studies, thematic
studies and cost-benefit analysis
will complement efforts to assess
attribution.

4. Revise and strengthen the
contribution narrative: This is a
continuous process of testing and
revising the theory of change that
underpins the central argument
that GEMS is making a difference.
In this way contribution analysis has
a formative effect in that it allows
programmes to quickly understand
whether or not resources are being
used in an optimal way to deliver the
changes envisaged at the outset.
Based on this process, programmes
should assign a percentage to the level
of attribution that can be assigned
to observed changes in indicators
– for each level in the results chain.
Programmes should try to balance
accuracy and simplicity in doing this,
and thereby provide estimates which
are credible both within and outside
the GEMS, and at the same time are
manageable for staff to implement.
It is important that significant effort is
exerted in avoiding ‘self-importance
bias’ in the assignment of a percentage
to the level of attribution – applying
the 3-Step process and a combination
of quantitative and qualitative
measurement techniques.

• How difficult is it to evidence
these effects and why?

17. DCED (2010).
18. Much of this section is drawn from GEMS2 (2012).
19. Outcome harvesting is one of a range of techniques that may be helpful in identifying explanations of change.
20. Note that some programmes (for example improving the business environment) create pre-conditions for development outcomes, rather than stimulating
actual change. Attribution (and measurement of impact) may be more difficult in such cases.


7

DATA SOURCES
AND COLLECTION
METHODOLOGIES

30

31


7

7. Data sources and collection methodologies

DATA SOURCES AND COLLECTION
METHODOLOGIES

7.1. Data sources
A mixed method approach to results
measurement implies the use of a
combination of quantitative and
qualitative data sources – both drawn
from intervention-specific and sector/
programme-wide surveys. The selection
of data sources should be included
in component-wide and interventionspecific measurement plans.
Wherever possible, impact and
outcomes should be measured using
quantitative measurement of the
target population. In following the
3-Step approach to measurement,
quantitative data will be required of:
• Changes at the intervention level
and, ideally, a counterfactual
comparison.
• Wider surveys of the target
population (e.g. through the crossGEMS Enterprise Survey).
In most cases, rigorous estimates of
attribution will require qualitative
methods as a complement to
quantitative measurement. This will
help to triangulate evidence of impact
and test the plausibility of intervention
pathways specified in the GEMS
results chains. Qualitative approaches
will also provide a mechanism for
identifying unforeseen consequences.
This is very important when impacts
are dependent on complex pathways
of change, including crowding-in and
catalyst effects, since the definition of
target indicators relies on assumptions
about market sensitivity to small
events. Qualitative approaches are also
valuable in determining the reasons
that particular types of impact were or
were not realised.

21. World Bank (2011).

Many GEMS interventions will go
through an initial piloting stage.
Rigorous measurement at this stage in
the project will often be valuable for
three reasons:
• A pilot will normally be more
‘contained’ than subsequent roll-out
of the intervention, meaning that it
will be easier and more cost effective
to identify a comparison group and
isolate attributable change.
• The data gathered at this stage in an
intervention is valuable to assess its
viability, learn lessons in regarding
optimal intervention design, and
to accordingly adjust the design of
the intervention in any subsequent
scaling up.
• Estimates of the scale of impact
derived during a pilot stage may
be used as proxies for estimating
the scale of change at the scaling
up stage – where it will be more
challenging and costly to measure
change and identify a comparison
group. Where this is done, a
methodology should be put in
place to regularly validate the
extrapolation when changes in
indicators for large numbers of
enterprises are calculated using data
from small samples or a pilot phase.

7.2. Data collection
strategy
GEMS components should develop a
data collection strategy that answers
the following questions:
• At what point in time should the
impact be measured?
• What data are required?
• What is already being collected/
available?
• What additional data need to be
collected?
• Who will be responsible for data
collection and what processes need
to be set up?
The strategy should set out the
design of the results measurement
activities, including surveys, qualitative
methods to be used, the timing of
these activities, and other aspects of
results measurement design. Indicators
are required throughout the results
chain, meaning that data collection
requirements are likely to include:
• Administrative data on the delivery
of the intervention. At a minimum,
monitoring data are needed to
know when a programme starts
and who receives benefits, as well
as to provide a measure of the
‘intensity’ of the intervention in cases
when it may not be delivered to all
beneficiaries with the same content,
quality, or duration.21

7. Data sources and collection methodologies

• Survey data: GEMS components will
require a combination of:
> Bottom-up surveys, which are
intervention-specific to address
the Step 1 evaluation questions.
These will be tailored to each
specific intervention and will
generally focus on the direct
beneficiaries of the intervention
to assess the degree to which
intended outputs (and outcomes)
have been achieved.
> Top-down surveys which will
cover the sector as a whole to
address the Step 2 research
questions. For most components,
this will be undertaken through
the cross-GEMS Enterprise
Survey. For GEMS 2, a separate
construction and real estate
sector survey will be undertaken.
Surveys should collect information
on a sample of enterprises in an
area where the project expects to
work, as well as on a comparable
sample in an area in which no
project activity is planned. GEMS
components should closely
supervise data collection to
ensure a good quality dataset.
Following the baseline, since
a set of panel data is desired,
data collected from particular
respondents during the different
survey rounds must be consistent
with each other.
• Qualitative research: This research
can take the form of interviews, focus
group discussions, and in-depth
histories/case studies. A variety
of methods may be appropriate,
including outcome harvesting,
most significant change and
qualitative comparative analysis.
Some qualitative research may
be incorporated into the survey
work. It is likely that this will be
complemented by in-depth
consultations undertaken by GEMS
intervention managers.
• Data on exogenous factors that
may affect the outcome of interest.
These make it possible to control for
outside influences.
• Secondary data and other
information, including sector
analysis and reports, household
surveys, etc.

7.3. Survey guidance
Generic guidance on good practice
in collecting baseline information and
undertaking research is provided in
Section 3 of DCED (2010). Research
should be in line with established good
practices for choice of data gathering
tools, planning, questionnaire (or other
instrument design), sampling, data
gathering, supervision, data entry,
analysis and research management. Ten
criteria for a good survey are provided
in Box 6.
GEMS faces significant challenges
in ensuring that data collected
through surveys or other means is
‘representative’22. This is largely due
to the following characteristics of the
programme:
• The ‘target population’ is in many
cases very large. For example, it is
not possible to isolate the potential
beneficiaries of GEMS 3 in any given
state: it has the potential to deliver
income and employment benefits
to a wide cross-section of the
population.
• In many cases, data inadequacies in
Nigeria and a large and dominant
informal sector mean that total
population figures from which
a robust sampling frame can be
derived are not available.
• The degree of market-level change
that can be facilitated by GEMS is
likely to be relatively small relative
to the wide range of explanatory
variables that exist.
• High levels of attrition are likely
in many cases, meaning that the
sample size will be considerably
larger than would otherwise be
required.


This handbook does not set fixed
guidelines for sample sizes, margins
of error and confidence levels. GEMS
must be transparent in the methods
used for measurement and report
the confidence levels and margins
of error of the quantitative results
reported. Given the need to ensure
that the cost of measurement is
proportionate, a mixed method
approach that triangulates results will in
most cases be appropriate, providing
a convincing body of evidence, whilst
acknowledging uncertainties in the
accuracy of the reported results.

7.4. Measurement
frequency
The GEMS logframe defines indicators
on an annual basis during the period of
implementation, plus a final indicator
two years after implementation ends
– to take account of time taken to
achieve results from systemic change
and to assess sustainability. It may not
be cost effective to undertake a full
results measurement process on an
annual basis, although this is likely to
vary depending on the characteristics
of individual GEMS components and
interventions. As a minimum, GEMS
components should report on output
indicators on an annual basis and on
outcome and impact indicators at the
baseline, mid- and end-point of the
programme, and plan for an additional
survey two years after the completion
of the programme23. Responsibility for
completing the final survey still needs
to be agreed.

Box 6: Ten criteria for a good survey
1. The target population is well defined
2. The sample matches the target population
3. The sample is randomly selected
4. The sample size is large enough
5. Good follow-up minimizes non-response
6. The type of survey is appropriate
7. The questions are well worded
8. The survey is properly timed
9. The survey personnel are well trained
10. The survey answers the original question
Source: DCED (2010)

22. Saying that survey data is ‘representative’ normally implies that it can be said with 95% certainty that the observations reported from a selected sample are
the same as would be for the entire ‘population’.
23. It may be possible to report preliminary estimates on impact and outcome level indicators on a more regular basis. However, it should be made clear where
reported figures have not followed the 3-Step approach and may therefore not take account of deadweight loss and displacement.

32

33


ANNEXES

Annexes

1


GEMS LOGFRAME

GEMS aggregated logframe (impact and outcome levels)
Impact: To increase growth, income and employment, especially for poor men and women, in target markets in
selected states and nationally
Indicator 1: Income (outreach)

Baseline
2011

Milestone 1
2012

Milestone 2
2013

Milestone 3
2014

Target 2015

Target 2017

Number of people recording
positive change in incomes

0

1,200

26,862

73,385

124,684

171,182

420

12,464

33,625

56,324

78,590

6,419

19,094

31,496

43,700

Notes: data to be disaggregated
by formal/informal. Average total
annual income increase to reach
15% or more

Number of poor
0

Number of female
0

36

Source
Programme/intervention monitoring information; surveys of representative sample of target and
control group firms and enterprises
Indicator 2: Income (value)

Baseline
2011

Milestone 1
2012

Milestone 2
2013

Milestone 3
2014

Target 2015

Target 2017

Aggregated change in
cumulative income (GBP)

0

£600,000

£11,036,610

£26,771,429

£45,267,402

£64,101,775

£210,000

£5,646,810

£13,286,599

£21,866,757

£32,000,747

£-

£1,269,277

£3,722,352

£6,087,054

£8,536,313

Notes: to be further
disaggregated by formal/informal

Poor
0
Female
0
Source

Indicator 3: Employment

Baseline
2011

Milestone 1
2012

Milestone 2
2013

Milestone 3
2014

Target 2015

Target 2017

Change in employment
(FTE jobs)

0

0

2,795

14,434

24,080

29,776

1,533

4,468

7,297

10,259

Notes: to be further
disaggregated by: jobs created/
people into employment; formal/
informal.

Number of females
0

0

Source

34

35


Annexes

Impact: To increase growth, income and employment, especially for poor men and women, in target markets in
selected states and nationally
Indicator 4: Firm growth
(outreach)

Baseline
2011

Milestone 1
2012

Milestone 2
2013

Milestone 3
2014

Target 2015

Target 2017

Number of firms (including self
employed) with increased sales

0

1,000

11,133

25,859

45,523

62,727

Notes: disaggregated by: male/
female managed or owned firms;
formal/informal

Source

Indicator 5: Firm growth (value)

Baseline
2011

Milestone 1
2012

Milestone 2
2013

Milestone 3
2014

Target 2015

Target 2017

Increase in sales amongst
targeted firms

£-

£2,200,000

£111,500,000

£228,500,000

£356,500,000

£568,500,000

Notes: disaggregated by: male/
female managed or owned firms;
formal/informal. Target currently
excludes GEMS 3 –TBD

Source

Indicator 6a: Systemic change
and sustainability: private
sector

Baseline
2011

Milestone 1
2012

Milestone 2
2013

Milestone 3
2014

Target 2015

Target 2017

Percentage of new or improved
products and services,
introduced through project
facilitation, that are established
in the market 12 months after
project support has ended

0%

0%

28%

43%

55%

62%

Indicator 6b: Systemic change
and sustainability: public sector
and civil society

Baseline
2011

Milestone 1
2012

Milestone 2
2013

Milestone 3
2014

Target 2015

Target 2017

Percentage of new or improved
regulations or reforms,
introduced through project
facilitation, that are established
in the market 12 months after
project support has ended

0%

0%

28%

43%

55%

62%

GEMS 3 only Additional
outcome indicator D: Time
and cost of Doing Business in
Nigeria

Baseline
2011

Milestone 1
2012

Milestone 2
2013

Milestone 3
2014

Target 2015

Target 2017

Percentage improvement
in Nigeria’s absolute Doing
Business rating

100

99

96

92

90

86

Notes: indexed by GEMS 3;
baseline = 100

Doing Business dataset

Indicator 7: Product quality

Baseline
2011

Milestone 2
2013

Milestone 3
2014

Target 2015

Target 2017

Improvements in product
quality (variously defined by
component)

Target non aggregatable - see separate component logframes for target details

Source
Note: Enterprise surveys, intervention specific evaluation reports

Source
Note: Enterprise surveys, intervention specific evaluation reports

Source

Milestone 1
2012

Source
Various annual consumer perception surveys

Annexes

2


GLOSSARY OF TERMS

Attribution

Attribution is defined by Glossary of Key Terms developed by the DAC Network on
Development Evaluation as the ascription of a causal link between observed (or expected to be
observed) changes and a specific intervention.24 In simple terms this means that attribution refers
to extent of change that can be claimed by a project/intervention out of the total change that
takes place.

Counterfactual

What would have happened if the intervention had not taken place.

Crowding in25

Enterprises at levels other than the target level copying behaviours that those affected by
programme activities have adopted; or entering a sector or value chain as a result of improved
incentives and environment created (at least partly) by the programme. This term also applies
to government agencies or civil society organizations, who are not directly involved in the
programme, copying behaviours of those who are directly involved in the programme, or who
change their behaviour as a result of improved incentives or environment created (at least partly)
by the programme.

Deadweight loss

The proportion of total results that would have been secured anyway in the absence of the
intervention.

Displacement

Some enterprises may be negatively affected because others are benefiting from programme
activities. Displacement is the amount of negative effect on those enterprises harmed by
programme activities.26

Full time equivalent

The ratio of the total number of hours worked to what would be undertaken by a full time worker.
A full time worker is defined as someone who works for 8 hours per day, 240 days per year.

Informal employment

Includes both what is defined by the International Conference of Labour Statisticians (ICLS) as
‘employment in the informal sector’ and ‘informal sector employment’. Detailed definitions are
available at http://ilo.org/public/english/bureau/stat/download/papers/def.pdf

Intervention

An activity or series of activities designed to achieve a specific change in the support functions/
market or a set of activities designed to achieve the sustainable delivery of a new or improved
service or output that, through its use by the target group, will result in improved business
environment and contribute to increased incomes for that target group.

Most significant
change

Most significant change is a form of participatory M&E that involves project stakeholders both in
deciding the sorts of changes to be recorded and in analysing the data collected. The process
involves the collection of significant change stories emanating from the field level, and the
systematic selection of the most significant of these stories by panels of designated stakeholders
or staff.

Multiplier

Further economic activity (e.g. jobs, expenditure or income) associated with additional income
to those employed as a result of the intervention (income multipliers), with local supplier
purchases (supplier multipliers) and with longer term development effects (dynamic effects e.g.
induced inward migration).

Sustainable
intervention

An intervention that remains established in the market 12 months after project support has
ended.

Systemic change

Changes in market systems and the structures, such as government and civil society, that support
markets that cause sustainable shifts in the way those market systems and structures operate,
for example, changes in relationships within and among both private enterprises and public
agencies, in incentives and in market support structures. Systemic change causes widespread
indirect results such as crowding in, copying, enterprises shifting sectors and changes in
enterprise start-up and exit rates.27

24. http://www.oecd.org/dac/2754804.pdf
25. Source: DCED.
26. Ibid.
27. Ibid.

36

37


3

Annexes

REFERENCES AND FURTHER READING

1. Caputo, E. with A. Lawson, A. de Kemp, S. Leiderer and C. Valmarana (2011),
Application of new approach to the evaluation of Budget Support operations:
Findings from Mali, Zambia and Tunisia – December 2011.
http://www.oecd.org/countries/mali/49716695.pdf
2. DCED (2010), Measuring Achievements in Private Sector Development:
Implementation Guidelines, Version 1g, 5th March 2010.
http://www.value-chains.org/dyn/bds/docs/702/Implementation%20Guidelines_
DCED_StandardVersionV.pdf
3. DFID (2012a), Measuring job creation 2: How do we define a job?, How to note,
June.
4. DFID (2012b), Broadening the range of designs and methods for impact
evaluations, Report of a study commissioned by the Department for International
Development, Working Paper 38, April.
http://www.dfid.gov.uk/Documents/publications1/design-method-impact-eval.pdf
5. GEMS1 (2012), Monitoring and Results Measurement Strategy and Plan for
GEMS1, 29 February.
6. GEMS2 (2012), M&E Framework, Version 5, 13 July.
7. GEMS3 (2012), Monitoring and Evaluation Plan, April.
8. IMEP (2012), GEMS Monitoring and Evaluation System Quality Review, 26 April.
9. London Development Agency (2010), 2010 Metrics Handbook: A guide to the
LDA’s outcomes, intermediate outcomes and outputs.
10. NONIE (2009), Impact Evaluations and Development, NONIE Guidance on Impact
Evaluation.
http://siteresources.worldbank.org/EXTOED/Resources/nonie_guidance.pdf
11. White, H. and Phillips, D. (2012), Addressing attribution of cause and effect in
small impact evaluations: towards an integrated framework, 3ie Working
Paper 15.
http://www.3ieimpact.org/media/filer/2012/06/29/working_paper_15.pdf
12. World Bank (2011), Impact Evaluation in Practice.
http://siteresources.worldbank.org/EXTHDOFFICE/
Resources/5485726-1295455628620/Impact_Evaluation_in_Practice.pdf
13. World Bank (2012), Nigeria: poverty trends between 2004 and 2010:
methodological review and an update, A Background Note, August.

Annexes

4


NIGERIA HOUSEHOLD COMPOSITION
BY ZONE (2011)

Nigeria household composition by zone (2011)
Zone

Nigeria

NC

NE

NW

SE

SS

SW

Less than 15

43.6

50.3

51.7

34.3

39.2

35

43.2

15 to 64

52.8

46.4

45.3

56.2

56.4

58.5

52

65 and older

3.7

3.4

3

9.5

4.4

6.5

4.8

Average household size

6.04

7.42

6.75

4.57

5.44

4.4

5.63

Net income per
household required to
keep household above
poverty line (₦)

497,720

611,438

556,227

376,586

448,278

362,578

463,935

Net income per worker
required to keep
household above poverty
line (₦)

156,068

177,595

181,907

146,626

146,106

140,862

158,469

Household composition
by age group (%)

Source: Derived from General Household Survey (GHS) conducted in 2010/2011 by the National Bureau of Statistics.28

28. The GHS was a detailed survey administered to approximately 5,000 households comprising 28,000 individuals. It is representative at the national, sectoral
(urban/rural), and zonal levels.

38

39


5

Annexes

INTERVENTION-SPECIFIC GUIDANCE

The DCED Standard requires that
results chains are developed for
every intervention to show how
the intervention will lead to the
achievement of development goals and
to mention other relevant contributions.
This has been done by all GEMS
components.
To provide guidance on the application
of the measurement approach
described in the main body of this
handbook, this section presents
results chains for five common
categories of GEMS intervention and
provides suggested data sources and
measurement methods for each. The
examples provided are illustrative and,
as a result, are generic and simplistic.
They are not representative of every
GEMS intervention and will need to be
adapted to reflect specific contexts.

The five categories are:
• Support to BMOs and related
advocacy and industry coordination.
• Skills development and capacity
building.
• Establishment of new or improved
products, services or processes.
• Business enabling environment
reform.
• Business support services.

Based on information provided by the
four GEMS components, all existing
or planned interventions have been
listed and categorised according to
the five categories listed above (see
Annex 6). As is inevitable in a process
of developing generic categories for
a large set of disparate interventions,
some fit more neatly into these
categories than others.

Annexes


Support to BMOs and related advocacy and industry coordination
Figure 6: Intervention results chain
Impact
(poverty level)

Income
(value and
outreach)

STEP 2

Employment

Impact
(sector level)

Systems of
representation and
advocacy functioning
better

EQ2.1

Firms perform
better and grow

STEP 2

Impact
(enterprise level)

Improved
business
environment

EQ2.2

STEP 3

Outcome
(enterprise/core
market)

STEP 1

Output
(service/support
market)

STEP 2

EQ3

Policy changes as a
result of advocacy

EQ1.3
Firms and other
stakeholders use
or benefit from
BMO services

EQ1.2

EQ1.3

EQ1.2
BMO is
established

BMO grows capacity
and performs better

EQ2.3
EQ1.1
Intervention
Support to BMO to become established and
improve performance

Table 7: Summary of suggested data sources and measurement methods – support to BMOs and related advocacy
and industry coordination
Data sources
Step 1: Intervention to market
system change (support and
core markets)

Measurement methods

• Monitoring data.
• Key informant interviews and focus group

• Institutional capacity assessments.
• Political economy analysis.
• Qualitative analysis and triangulation.
• Quantitative analysis of survey results.

discussions.

• Survey of users of BMO services.
Step 2: Changes in income and
employment in core market

• Household survey and HNLSS.
• National accounts.
• GEMS Enterprise Survey.
• Relevant reports.

• Analysis of relevant data and documentation.
• Regression or difference-in-difference analysis.

• Findings from Steps 1 and 2.

• Synthesis and triangulation.
• Qualitative methods, including Perception

discussions.

Step 3: Compare and
triangulate findings – linking
core market change to GEMS

Surveys, Most Significant Change analysis.

40

41

Research question

Judgement criteria

measurement

Implications for measurement
design and methods

• Existence of BMO.
• Organizational capacity

• What other factors are driving the

• Key informant interviews and

• Member perceptions of BMO

• Is there a baseline capturing initial

Suggested indicators

Step 1: Intervention to market system change (support and core markets)
EQ1.1

What have been the direct outputs
from GEMS interventions?

• To what extent have GEMS
interventions led to:

> The establishment of the BMO.
> Improved BMO capacity and
strengthened performance.

assessment of BMO.
performance.

establishment of the BMO and/or
its improved capacity?
levels of capacity in the BMO?

What changes have occurred at
the service/support market level?

other stakeholders increased
the use of BMO services and
benefited from this use?

• To that extent has GEMS enabled
firms to benefit from BMO
services?

• BMO membership numbers.
• Number of private public

• Is there a baseline for indicators?

dialogues held.

• Firm level capacity and
performance.

• Firm growth (value).

• ‘Before and after’ survey of users
of BMOs (i.e. members and
companies using their services).

• BMO administrative data.
• Can we define the pathways

through which GEMS support
will lead to benefits from BMO
services?

• Are the BMOs pre-selected or
self-selecting?

• Is it possible to define a


focus group discussions with firms
using BMO services.

• Survey of users of BMOs (i.e.

members and companies using
their services) – both for treatment
and comparison groups.

comparison group of firms with
matching characteristics which
do not use BMO services? What
are the key matching variables
or criteria to define appropriate
comparators?

• Is there likely to be spill over

and contamination between the
treatment and comparison group?

What have been the other causal
factors?

• What other factors have led to:
> The establishment of the BMO.
> Improved BMO capacity and
strengthened performance.
> The use of BMO services by
firms and other stakeholders.

• To what extent are there other


• To what extent are there likely

• Survey of users of BMOs (i.e.

‘contributory causes’ to the
achievement of the results?

to be multiple benefits or
‘unintended consequences’?

• How significant is the intervention,
in terms of the identifiable change
it represents?

focus group discussions with firms
using BMO services.
members and companies using
their services) – both for treatment
and comparison groups.

Annexes

To what extent can these changes
be attributed to GEMS?

• To what extent have firms and

• ‘Before and after’ institutional

capacity assessments of BMO.

• Completion of business plans.
EQ1.2

focus group discussions with
BMOs receiving support.


Table 8: Measurement design and methods selection framework – support to BMOs and related advocacy and industry coordination


Research question
EQ1.3

Judgement criteria


measurement

design and methods

To what extent have the service/
performance and inclusiveness of
the core market system?

• To what extent have BMO

• Number of policy changes as a

• Political economy analysis


activities had an effect on:
> Policy changes as a result of
advocacy.
> Improvements in the business
environment.

result of advocacy.

• Business environment indicators.
• Number of BMOs replicating

required to understand drivers
of policy reform and the
effectiveness of advocacy.

focus group discussions with:
firms using BMO services; BMOs;
policy makers.


beneficiary BMO.

• What have been the other
contributory factors?

• Can we observe any systemic

changes in the market system
for advocacy beyond the
immediate results from the GEMS
intervention?

Step 2: Changes in income and employment in core market
EQ2.1

• Income and employment should,

What market-level factors have
led to the changes in income and
employment?

• What have been the key

if possible, be measured at the
household and enterprise level.

determinants of:

> Improved firm performance
and growth.

> To what extent have
improvements in policies and
the business environment been
a contributory factor?


• Sources of income and

• What sample size is required for it

• GEMS Enterprise Survey to

• Systemic change and

• To what extent are secondary

• Regression or difference-

employment.

sustainability.

• Key determinants of changes to
firm performance.

characteristics of the population –
to enable matching etc?

to be representative at a national
and/or state level?

data sources available, timely and
reliable?

• General Household Survey and
HNLSS (although infrequent).


establish improved growth and
firm performance.
in-difference analysis of
determinants of sector growth.
Comparative analysis with other
states where GEMS did not
provide support.


focus group discussions to
establish causes of improved firm
performance and importance
of policies and business
environment.

• Analysis of relevant data and
documents.


> Growth in incomes and
employment.

• Average incomes.
• Employment levels and wages.

Annexes

EQ2.2

How have the level and
relevant sector/geographic area)?

42

43

Research question
EQ2.3

Judgement criteria

What have been the causes of any
changes in market conditions?


measurement

• What have been the causes of the

design and methods

• Doing Business Survey.

changes identified in EQ2.2?

• To what extent have BMOs had

an influence on the determinants
of improved firm performance
and growth in incomes and
employment?

environment.




documents.

EQ3.1

To what extent can the market
interventions?

• To what extent have the causes of
improved firm performance and
income and employment growth
been affected by the outputs of
GEMS support to BMOs?

• Time lag between changes in

BMO performance feeding into
improved firm performance
and growth in income and
employment.

• Significance of GEMS

interventions in the wider context
of business representation and
advocacy in the sector.

• Ensure that deadweight loss

and displacement are taken into
account in triangulation exercise.

• Synthesis and triangulation of
analysis from Steps 1 and 2.

• Qualitative methods, including
Perception Surveys and Most
Significant Change analysis.

Annexes

Step 3: Compare and triangulate findings – linking core market change to GEMS

Annexes


Skills development and capacity building
Impact
(poverty level)

Income
(value and
outreach)

STEP 2

Employment

Impact
(sector level)

EQ2.1

Firms perform
better and grow

STEP 2

Impact
(enterprise level)

Growth and
development of
the sector

EQ1.3

EQ2.2

Firms more
effective
and better
meet clients
requirements

STEP 3

Outcome
(enterprise/core
market)

EQ1.3

Workers possess
better skills

STEP 2

Output
(service/support
market)

STEP 1

EQ3
Strengthened
capacity of
organizations to
provide training

EQ1.2

Enhanced
training and skills
development
provision

EQ2.3
EQ1.1

Intervention

Support to organizations in training and skills
development

Table 9: Summary of suggested data sources and measurement methods – skills development and capacity building
Data sources

Measurement methods

Step 1: Intervention to
market system change
(support and core
markets)

• Survey of direct beneficiaries (training institutions and firms).
• Interviews, consultations, focus group discussions.

• Market analysis.
• Cohort tracking study.
• Quantitative analysis of survey data.

Step 2: Changes
in income and
employment in core
market

• Key informant interviews and focus group discussions.




Step 3: Compare and
triangulate findings
– linking core market
change to GEMS

documentation.

• Regression or difference-in-difference
analysis.


44

45

Research question

Judgement criteria


measurement

design and methods

• Are the direct beneficiaries

• Review of monitoring data

EQ1.1

Examples of outputs 29

• Co-operation agreements

between key market players
for provision of off the job skills
training


> Strengthened capacity by
vocational training institutes to
deliver training.

• Number of people trained.
• Satisfaction of trainees.
• Results from training courses
(where applicable).

> Better capacity of firms (formal
and micro) to provide training.

• Capacity of key market players
• Strengthening of capacity of

vocational training institutions
to provide training (trainers of
trainers, new training programmes
etc

• Vocational training institutions are
delivering training

• Strengthened capacity on part of
firms (formal and micro level) to
provide on the job training.

specifying numbers of direct
beneficiaries (training providers
and firms) of the programme

• Survey of direct beneficiaries

levels of capacity and provision
of training to this sector and by
whom?

(training institutions/firms) to
measures changes in supply side
provision (quality and quantity) in
that particular sector, before and
after GEMS.

• Does the rationale for GEMS in

• Counterfactual – survey of market

• Is there a baseline capturing initial

the intervention design provide
the basis for intervention in the
identified areas of intervention –
based on skills gaps, growth areas
in the sector etc.

• Is it possible to define appropriate
comparators in the training/
skills market amongst the key
market players30 that have not
been supported by GEMS. What
are the key matching variables
or criteria to define appropriate
comparators?

players not supported by GEMS
to examine their training provision
and how has it been changing
over the period covered by
GEMS. This should aim to isolate
differences between what market
players not supported by GEMS
are doing versus what market
players supported by GEMS are
delivering. It should also help
establish market (training market)
effects on them as a result of
GEMS intervention.

• Interviews, consultations, focus

group discussions with the
training infrastructure (e.g.
providers and firms) and final
beneficiaries of training support
(individuals and firms) to explore
how and in what way GEMS is
influencing the skills/training
provision for the supported
sectors.

29. Source: results frameworks for GEMS 2 – revised April 2012.
30. BMOs, training organizations, other institutions and individual trainers are defined as the key market players.

Annexes

increased to provide off the job
training

identifiable and known – e.g.
vocational training institutes, firms
providing training (both formal
construction firms and micro level
firms).


Table 10: Measurement design and methods selection framework – skills development and capacity building


Research question
EQ1.2

Judgement criteria


measurement

design and methods


• To what extent has there been an

• Number of training courses in

• How is the market structured – are

• Interviews and consultations with

overall improvement in training
provision?

• Are the changes to the training/

existence.

there different segments, different
organisations active in those
segments?

• Overall market analysis of the

skills market system wide?

training /skills provision for the
sector (supply and demand sides).

• Has improved training provision
led to an increase in skills in the
workforce?


• What changes have been made

to training programmes as a result
to GEMS?

• How has the quality and quantity
of the supply of skilled trainers
to the sector been affected by
GEMS?

key market players to appreciate
market changes and factors
driving change.

• Can we define the treatment and
the change pathways through
which results are expected to be
attained?

• Are there differential effects on

different market segments – can
we break up the market into
segments that are more affected
by GEMS and those less affected?

• Can we compare the quality of

• Construction of a comparison

• Survey of direct beneficiaries and

a comparison group to examine
changes in quality and quantity in
training provision.

• Analysis of the market for training
provision to sector.

• Feedback from end beneficiaries
(individuals and their employers)
on the quality of training
provision – cohort tracking
study of final beneficiaries/end
users taking a representative
sample of individuals trained
and interviewing them as well as
their employers to establish how
skills provision has changed and
how it is feeding into business
performance changes.

key market players.


group may be compromised
by possible spillovers and
contamination of comparators.
Non GEMs providers may
respond to market competition
by offering new/better courses
similar to those provided by those
supported by GEMS providers.

beneficiaries (firms, training
providers) capturing the numbers
of new trainers, changes in their
capacity levels, numbers of new/
enhanced courses designed/on
offer, additional qualifications on
offer, etc.

Annexes

provision on offer by firms and
training providers as a result
of GEMS versus the quality of
training provision by those firms/
training providers not supported
by GEMS – Is it possible to
construct a comparison group?

• Monitoring data on direct

46

47

Judgement criteria

factors?

changes in the training/skills
development market for that
sector .

• Trend analysis of demand by

• To what extent are there likely to

• What have been the wider

design and methods



measurement

• Overall market analysis of the

• How significant is the GEMS

Research question


• Are the end users or final

• Feedback from end beneficiaries

contributory causes leading to the
achievement of the results
be multiple benefits/unintended
consequences

intervention in terms of the
identifiable change it represents.

EQ1.3

• To what extent has improved

training and skills provision in the
sector fed into improved business
performance?

• Performance of firms providing

training or where trained workers
are employed.

beneficiaries known or identifiable
e.g. the individuals trained as a
result of GEMS intervention? Also
are their employers identifiable?

• Is it possible to break down the

changes in the performance of the
core market as a result of training
targeted at different groups e.g.
youth, apprentices, unemployed,
employed apprentices, artisans
etc.

• What is the appropriate time lag

between skills improvements
and improvements in business
performance? How does this vary
between firms in the sector?

sector for skilled workers and
unemployment levels
training/skills provision for the
sector (supply and demand sides)
key market players

(individuals and their employers)
on the quality of training
provision – cohort tracking
study taking a representative
sample of individuals trained
and interviewing them as well as
their employers to establish how
skills provision has changed and
how it is feeding into business
performance changes and plans
by their employers to hire more
workers.

• Feedback from a comparison

group – not benefitting
from GEMS (individuals and
enterprises) – what have been
the changes to their business
performance? Are there any
spillovers from GEMS impacting
on their performance.

EQ2.1

relevant sector/ geographic area)?

• What changes in income and

employment can we identify in the
sector/geographic region in which
skills interventions have been
focused?


• Is it possible to obtain reliable

• General Household survey and

• Can we identify a comparison

• Alternative data sources likely

and up to date information from
secondary sources, with the
requisite levels of disaggregation?
group?

HNLSS provide estimates of
income and employment. But
infrequent.

to be required, including GEMS
Enterprise Survey.

• Labour market data/surveys

capturing trends in incomes and
employment levels in sector

Annexes





Research question
EQ2.2

Judgement criteria


measurement

design and methods

employment?


• Source of incomes and

• Time lag between changes in

• Regression analysis of the

• Improved firm performance and


determinants of:
growth.

• Growth in incomes and
employment.

• What is the strength of the

relationship between income
levels and skills in the sector?

employment.

sustainability.

firm performance.

• Performance of skilled workers
and firms employing trainees.

training provision feeding into
improved firm performance
and growth in income and
employment

determinants of changes in
income and employment, which
includes skills as a dependent
variable.


interventions in the wider context
of the sector and access to skilled
workers by the sector

• A multitude of other factors

affecting improvements in firm
performance other than skills
related issues

• Is it possible to break the

sector into segments based on
expectations that some parts are
more and others less affected by
skills / training type interventions?



• What are the sector level drivers of

• What have been the causes

EQ2.3

• Is GEMS targeted at the key


drivers of growth and performance
through its interventions – e.g.
appropriately targeted at skills
gaps, areas with growth potential.

• GEMS Enterprise survey.
• Market analysis of training

provision (demand and supply).

• Interviews and consultations

with firms to establish plausible
attribution of GEMS to trends in
firm performance, income and
employment.

Annexes

of identified improvements in
workforce skills?

growth and performance?

interventions?

GEMS support in terms of skills
development and training?

• Taking the impact of GEMS

on the training market and the
data on the changes in business
performance, can we distil with
some degree of confidence what
elements of the changes in the
latter are due to GEMS. Can we
quantify the changes in income
and employment of poor people
due to GEMS?



• Triangulation of the findings

from Steps 1 and 2 to build the
evidence that the changes on
the training / skills market due to
GEMS have:
> plausibly resulted in changes
in business performance in the
core market which in turn have
> plausibly resulted in
quantifiable changes in levels
of employment and income for
poor people.


48



EQ3.1


Annexes

Establishment of new or improved products, services or processes
Income
(value and
outreach)

STEP 2

Impact
(poverty level)

Impact
(sector level)

Firms perform
better and grow

STEP 2

Impact
(enterprise level)

EQ1.3

EQ2.2

Firms produce to
a higher quality or
at a lower cost

EQ1.3

STEP 3

Outcome
(enterprise/core
market)

STEP 2

EQ3
Output
(service/support
market)

Employment

Growth and
development of
the sector

EQ2.1

STEP 1

49

EQ2.3
Intervention

Businesses have
understanding of
new or improved
products, services
or processes

EQ1.2

New or improved
products, services
or processes
adopted

EQ1.1
Capacity building or financial support to
businesses to introduce innovative products,
services or processes

Table 11: Summary of suggested data sources and measurement methods – establishment of new or improved
products, services or processes
Data sources

Measurement methods

(support and core
markets)

• Key informant interviews.
• Enterprise surveys.

• ‘Before and after’ studies.
• Comparisons with non-adopters of new

Step 2: Changes
in income and
employment in core
market

• National accounts
• Cross-GEMS Enterprise Survey.

• Outcome harvesting.
• Qualitative comparative analysis.


• Qualitative methods, including Perception Surveys

discussions.

Step 3: Compare and
change to GEMS

innovation.

and Most Significant Change analysis.

Table 12: Measurement design and methods selection framework – establishment of new or improved products, services or processes

Research question

Judgement criteria


measurement

design and methods

EQ1.1

EQ1.2



• Level of awareness of innovation.

• Require market information on




• Number of firms adopting

• Is it possible to identify

• Try to identify similar firms as a

interventions led to an improved
understanding of new/innovative
business models or production
techniques?
adoption of an improved product,
service or process.

improved the product, service or
process.

• Sales incorporating the improved

existing products and services
(baseline).

comparable firms that are not
applying the improved product,
service or process?

‘comparison group’.

• Sales and number of enterprises
involved.

product, service or process.


• To what extent has GEMS been

• Pathway leading innovation to

• Do target group perceive linkage

factors?

• What other factors have led to

• What else/who else is promoting

• ‘Before and after’ studies.

responsible for the improved

anticipated benefits must be
clear.

the adoption of the improved

innovations?

• What other market signals are

between intervention and
intended benefit?

influencing the decision to adopt
an innovation?
Annexes

• What other conditions need to be
in place for the innovation to be
successful?

EQ1.3

• To what extent has the improved

product, service or process led to
improved quality or lower costs?

• To what extent has this enhanced

• Product quality.
• Production costs.
• Firm sales and profitability.

firm performance and growth?

• What are the other influences

of quality, cost, and enterprise
performance?

• Comparison of firm performance
with firms not adopting new

• Is the market growing or
shrinking?

• What factors affect the speed of
adoption?

EQ2.1


• Income and employment should,
if possible, be measured at the
household and enterprise level.


• Is it possible to obtain reliable

and up to date information from
secondary sources, with the
requisite levels of disaggregation?

• Can we identify a comparison
group?





50

51

Research question
EQ2.2

Judgement criteria


measurement

design and methods

employment?


• Source of incomes and







determinants of:
growth.

employment.

• To what extent has adoption of

new product, service or process
been a contributory factor?

employment.

sustainability.


to be representative at a national
and/or state level?

reliable?

firm performance.


firm performance.

performance and importance of
new product, service or process.

• Performance of firms applying

new product, service or process.


Table 12: Measurement design and methods selection framework – establishment of new or improved products, services or processes

EQ2.3



the changes identified in EQ2.2?
To what extent have GEMS
interventions had an influence?

• What are the sector level drivers
of growth and performance?

• Is the sector growing?

documents.

EQ3.1

interventions?


been affected by the outputs
of GEMS support to business
models or production techniques?

• Can we distil with some degree of
confidence what elements of the
changes in business performance
are due to GEMS support to
improved products, services
or processes? Can we quantify
the changes in income and
employment of poor people due
to GEMS?




evidence that the improved
products, services or processes
due to GEMS have:
poor people.


Annexes

documents.

Annexes


Business enabling environment (BEE) reform
Income
(value and
outreach)

STEP 2

Impact
(poverty level)

Impact
(sector level)

Increased sales
from increased
competitiveness,
productivity,
quality

EQ2.1

Increased
transactions in
land, tax and
investment

STEP 2

Impact
(enterprise level)

EQ2.2

Change in business
practices: expansion/
start-up; A2F;
increased trade;
reduced vulnerability

EQ1.3

STEP 3

Outcome
(enterprise/core
market)

Employment

Improved BE
that responds to
enterprises needs

EQ1.3

Demonstrated
interest and
capacity of
government to
enhance BEE

STEP 2

Output
(service/support
market)

STEP 1

EQ3

EQ2.3
Intervention

EQ1.2

Improved services
and ‘products’
(policies, rules,
regulations) in land,
tax and investment

EQ1.1
Interventions to address BEE and constraints in
land, tax, investment systems

Table 13: Summary of suggested data sources and measurement methods – business enabling environment reform
Data sources

Measurement methods

(support and core
markets)

• Intervention monitoring data.
• World Bank.

• Comparative case studies.

Step 2: Changes
in income and
employment in core
market





Step 3: Compare and
change to GEMS

documentation.

• Regression or difference-in-difference
analysis.


Surveys and Most Significant Change
analysis.

52

53

Research question

Judgement criteria


measurement

design and methods

• Level of policy dialogue –

• What are the other factors driving


EQ1.1



> Enhanced interest and capacity
of government to undertake
policy and regulatory reform.

internally and externally (e.g.
number of meetings, etc).

• Publication of policy documents.

the changes in dialogue?

• Contribution of other GEMS
components.

> Initiation of evidence-based,
joined-up policy dialogue
EQ1.2


an improvement in the ‘products’
(policies, rules and regulations) in
land, tax and investment?

• To what extent has GEMS

supported the building of the
capacity of State Governments
to strengthen BE and introduce
reform?

• To what extent has GEMS

created/enhanced mechanisms
for dialogue and flow of
information between Government
and the private sector?

• New legislation/tax codes.
• Number of bureaucratic

• World Bank Doing Business

processes faced by business.

Indicators.


through which GEMS support will
lead to reform?

• Are beneficiaries pre-selected or
self-selecting?

comparison group?

• To what extent has GEMS worked
to change the capacity of the
private sector to engage in BE
reform?

factors?

• What are the factors affecting


• What are the factors influencing

• How significant is the intervention

State level capacity to engage in
BE reform?
the level of dialogue and
engagement between the
Government and private sector in
BE reform.

in terms of the identifiable
change it represents?


Annexes


• To what extent has there been


Table 14: Measurement design and methods selection framework – business enabling environment reform


Research question
EQ1.3

Judgement criteria


measurement

design and methods


• To what extent have regulatory

• World Bank Doing Business

• Is there a baseline concerning the


or policy changes at the state
level led to improvements in the
business environment?

• To what extent have these

changes led to increased
transactions in land, tax and
investment?

Indicators.

• Number of transactions in land,
tax, investment.

• Overall enterprise investment
levels.

market systems for land, tax and
investment?

• Is there a baseline capturing the

focus group discussions.

• Enterprise survey.

private sector’s views on the key
constraints to the conduct of
business?

EQ2.1

EQ2.2



employment?


employment can we identify in
the sector/geographic region in
which interventions have been
focused?
determinants of:

> Improved firm performance
and growth.
> Increased competitiveness,
productivity and quality.

> Has there been a change in
business practices, increased
trade, reduced vulnerability?








employment.

sustainability.

firm performance.


to be representative at a national/
State level?
reliable?


firm performance.
Comparative analysis with other
provide support.


documents.


environment.

Annexes

> Growth in incomes and
employment.


54

55

Research question
EQ2.3

Judgement criteria



measurement


design and methods




identified improvements in the
business environment?

environment.




documents.

interventions?

GEMS support relating to the
BEE.

• Time lag between the GEMS

support and changes to the
BEE and then another time lag
between changes in BEE and
feeding into improved firm
performance and income and
employment change.


interventions in terms of business
environment systems overall.




evidence that the improved
Business Enabling Environment
due to GEMS has:
poor people.


Annexes

EQ3.1

Annexes


Business support services
Income
(value and
outreach)

STEP 2

Impact
(poverty level)

Impact
(sector level)

Growth and
development of
the sector

EQ2.1

Firms grow and
become more
profitable

STEP 2

Impact
(enterprise level)

EQ2.2

EQ1.3

STEP 3

Outcome
(enterprise/core
market)

Employment

Supported firms
become more
efficient and/or
innovative

EQ1.3

STEP 2

Output
(service/support
market)

STEP 1

EQ3
Increased value
of and improved EQ1.2
supply of Business
Services

Firms use new or
improved Business
Support Services

EQ2.3
EQ1.1
Intervention

Supply and demand for different types of
Business Support Services supported

Table 15: Summary of suggested data sources and measurement methods – business support services
Data sources

Measurement methods

(support and core
markets)


• Quantitative analysis of survey results.
• Outcome Harvesting.
• Most significant change techniques.

Step 2: Changes
in income and
employment in core
market



Step 3: Compare and
change to GEMS



Surveys and Most Significant Change analysis.

56

57

measurement

design and methods


• Firms’ awareness of the existence

• What are the other factors

• Firms’ perceptions of the value


• Improved supply of Business

• Number of firms offering BSS.
• Variety of BSS provided.


• Value of BSS procured by firms.

• BSS should be defined in the


Judgement criteria

EQ1.1



Business Support Services.
Support Services.

EQ1.2


increased use of Business Support
Services by firms?

of BSS.

driving the changes in the factors
identified?

of BSS.

narrow context of the GEMS
intervention.

• Important to identify a

comparison group if possible.


• To what extent has GEMS support


been a contributory factor in the
increased use of new or improved
Business Support Services?

through which GEMS support will
lead to the use of BSS?

• Are beneficiaries pre-selected or
self-selecting?



comparison group?

factors?

• What other factors have affected


the market for Business Support
Services?


• How significant is the intervention

in terms of the identifiable change
it represents?

EQ1.3


• To what extent have changes

in the use of Business Support
Services made firms more efficient
and/or innovative?

• Have any changes in efficiency

and/or innovation led to
company growth and increased
profitability?

• Costs of production.
• Adoption of new technologies
or production/distribution
techniques.

• Firm sales and profitability.

• What are the other influences
of efficiency, innovation and
enterprise performance?

• Is the market growing or
shrinking?

• Comparison of firm performance
with firms not adopting new

Annexes


• Increased appreciation of value of

Research question


Table 16: Measurement design and methods selection framework – business support services


Research question

Judgement criteria

measurement

design and methods





EQ2.1

EQ2.2

relevant sector/ geographic area)?


employment?









employment can we identify in the
sector/geographic region in which
interventions have been focused?

determinants of:
growth.

• Increased competitiveness,
productivity and quality.


employment.

sustainability.



to be representative at a national/
State level?
reliable?

firm performance.

employment.

• Is there evidence of firms in the


firm performance.
Comparison analysis with other
provide support.


sector being more innovative and
adopting new technologies or
business practices?

environment.
documents.

EQ2.3



• What are the sector level drivers


• Is the sector growing?



environment.

documents.


identified innovations or adoption
of new technologies and business
practices?

of growth and performance?

Annexes


58

59

Research question

Judgement criteria


measurement

design and methods

• Time lag between the GEMS


EQ3.1

interventions?

been affected by the outputs
of GEMS support relating to
Business Support Services.

support and changes to the
provision of Business Support
Services and then another time
lag between changes in BSS
and feeding into improved firm
performance and income and
employment change.




poor people.


Annexes

interventions in terms of BSS
overall.

evidence that the improved BSS
due to GEMS has:



Annexes

6


CATEGORIZATION OF GEMS INTERVENTIONS
Support
to BMOs
and related
advocacy
and industry
coordination
GEMS 1

3

Skills
development
and capacity
building

Establishment
of new or
improved
products,
services or
processes

2

8

1

Feed Finished Animal for Sallah 2011

2

%

3

Improved abattoirs through Dentata

%

4

Creating better business linkages

5

Brand awareness and brand promotion

%

6

Improving skills of FLG actors through training

%

7

Formation of Regional Association Lapan

%

8

Formation of National Association

%

9

Demonstration of Policy lobbying through Lapan

%

10

Access to finance

%

11

Creating access to tools and equipment

%

12

Creating access to input (leather, sole, accessories etc)

%

13

Introduction of trade shows

Business
support
services

%

Feed Finished Animal Through Credit Programme

Business
enabling
environment
reform

%

%

GEMS 2

5

14

Advocacy and marketing support to BMOs to create
visibility with members and external stakeholders
and generate income

6

%

15

Provide technical support to bolster evidence base
and data about BMO members and their problems

%

16

Provide support to develop services for BMOs to
improve job opportunities for its artisan members,
assist in initial operations and facilitate linkages, e.g.
connecting them with construction companies

%

17

Support to the development of internal structures of
the LoCC

%

18

Development of internal and external
communication and advocacy capacity of the LoCC

%

19

Support formal vocational training institutions in
their institutional strengthening and organisational
development

%

20

Support specialised training organisations in
curriculum development, training aids and
communications for the CRES

%

21

Support BMOs, training organisations and other
institutions or individual trainers (key market players)
to develop and implement off the job skills training
for TBs

%

22

Support key market players to make agreements of
cooperation to develop and implement off the job
skills training for TBs

%

12

2

60

61


Annexes

Support
to BMOs
and related
advocacy
and industry
coordination

Skills
development
and capacity
building

Establishment
of new or
improved
products,
services or
processes

Business
enabling
environment
reform

Business
support
services

23

Support formal construction companies to develop
internal vocational training capacity

%

24

Support informal firms to develop internal vocational
training capacity

%

25

Create databases of artisans to feed into linkage
services from the builders artisans market

%

26

Identify and support service provider(s) to implement
and market the linkage services to stakeholders

%

27

Support service providers to facilitate training and
awareness workshop on contracts and negotiation
management

%

28

Support service provider to develop and implement
a communications strategy to build the B2B capacity
of artisans

%

29

Map the housing PPP landscape

%

30

Fill evidence gaps of effective demand and supply of
affordable housing

%

31

Facilitate roundtable meetings and actions plans
with PPP stakeholders

%

32

Facilitate availability of building technology suitable
for affordable housing

%

33

Implement capacity building programmes for MOH
stakeholders in response to the needs and gaps
identified in the roundtable action plans

%

34

Market research for specific input supplies for CRES

%

35

Technology transfer and piloting of viable input
supplies for CRES

%

36

Support commercialization and market uptake of
successful input supplies for CRES

%

37

Provide support to service providers to develop and
implement business models and business services in
the CRES

%

38

Support the development of demand within the
CRES for business management services

%

GENS 3

13

2

59

Federal
39

Formulation of Public Private Engagement
Mechanism (PPEM)

40

Development of Business Environment Improvement
Strategy (BEIS)

%

41

Application of a Nigerian Policy Framework for
Investment (based on the OECD PFI)

%

42

Sub-national Doing Business Survey

%

43

Prioritized Doing Business Action Points - Business
registration (including online business registration)

%

44

Establishing Public Private Engagement Mechanism
Advocacy, industry coordination and information
services

45

Support to Federal Minister of Finance Tax Policy
Unit Assistance

%

46

Support for a Policy dialogue between the (NERC)
and (NASME) on the implementation of the multiyear tariff order (MYTO)

%

%

%

Annexes

Support
to BMOs
and related
advocacy
and industry
coordination

Skills
development
and capacity
building


Establishment
of new or
improved
products,
services or
processes

Business
enabling
environment
reform

Kano
47

Mechanism (PPEM)

48

Strategy (BEIS)

%

49

Application of sub-national Policy Framework for

%

50


51

Training of Female Land Officers on sporadic land
registration

52

Workshop on Land Registration with information
officers from State MDAs and NOA officers
[delivered]

%

53

Land Regularization/Systematic Registration of Land

%

54

Investor Handbook [in the works: to be delivered
Mid-November]

%

55

Workshop on tax with treasurers and tax officers from
44 local government areas [14 November]

%

56

SOLA training related to land intervention [under
way for a week]

57

Assessing the level of need for support to GoK
[delivered]

%

58

MSME Strategy/Policy Support to MoCI and SEMT.
iPlan submitted for approval

%

59

Tax Harmonization/Tax for Service: iPlan in the worksready by 6th November

%

60

Informal sector/rural enterprise access to finance

%

61

Taxpayer information

%

%
%

%

%

Kaduna
62

Mechanism (PPEM)

63

Strategy (BEIS)

%

64


%

65


66

Tax harmonization

%

67

Integration of land registry systems and improving
public access to land information in Kaduna State

%

68

Kaduna Industrial and Finance Company
Improvement Strategy Work Plan

%

%

%

Cross River
69

Mechanism (PPEM) Advocacy, industry coordination
and information services.

70

Strategy (BEIS)

%
%

Business
support
services

62

63


Annexes

Support
to BMOs
and related
advocacy
and industry
coordination

Skills
development
and capacity
building

Establishment
of new or
improved
products,
services or
processes

Business
enabling
environment
reform

71


72

Advocacy, industry coordination and information
services.

73

Tax harmonization

%

74

Linkages for access to finance and infrastructural
support to Micro Enterprise Development Agency
(MEDA)

%

%
%

Lagos
75

Mechanism (PPEM)

76

Strategy (BEIS)

%

77


%

78

Prioritised Doing Business Action Points - Revenue
Administration Streamlining in Lagos Internal
Revenue Service (LIRS), as part of addressing ‘Doing
Business’ indicators

%

79

Development of a Strategic Urban Regeneration
Framework

%

80

Land Market Study

%

81

Partner with Lagos State Government in the
organization of Corporate Assembly and the
commemoration of Africa Industrialization Day

82

Reviewing and Updating of Accounting and Audit
Manuals for office Auditor-General for Local
Governments

%

83

Support to Lagos Inland Revenue Service (LIRS) on
Registration; Accounts; Cashless Arrangements; Data
Warehousing; Tax Audit tools and processed

%

84

Tax-for-Service/Improvement of Tax Harmonization
Implementation in Selected LGAs

%

85

Reviewing and Updating LGA Accounting and Audit
Manuals

%

86

Tax Harmonization Improvement Project in up to five
pilot LGAs

%

87

Improved Tax Complaints Process Project

%

88

Lagos Informal Sector Business Management
Development Program

%

89

Enterprise Support Units in selected businessoriented LGAs

%

90

Technical assistance for the establishment of an
Investment Promotion Agency (IPA)

%

91

Support to the Ministry of Women Affairs, Lagos
State on the development of Women’s Economic
Empowerment (WEE) Strategy/Policy

%

92

Development of Industrial and Competitiveness
Policy

%

%

%

Business
support
services

Annexes

Support
to BMOs
and related
advocacy
and industry
coordination

Skills
development
and capacity
building


Establishment
of new or
improved
products,
services or
processes

Business
enabling
environment
reform

93

Partnership with Manufacturers Association of
Nigeria in commissioning a Study on EU (Economic
Partnership Agreement-EPA) and ECOWAS
(Common External Tariffs) Protocols and their impact
on the manufacturing sector in Lagos State

94

Mechanism (PPEM)

95

Strategy (BEIS)

%

96


%

97

Prioritised Doing Business Action Points – Revenue
Administration Streamlining in Lagos Internal
Revenue Service (LIRS), as part of addressing ‘Doing
Business’ indicators

%

98

Development of a Strategic Urban Regeneration
Framework

%

99

Land Market Study

%

100

Partner with Lagos State Government in the
organization of Corporate Assembly and the
commemoration of Africa Industrialization Day

101

Reviewing and Updating of Accounting and Audit
Manuals for office Auditor-General for Local
Governments

%

102

Support to Lagos Inland Revenue Service (LIRS) on
Registration; Accounts; Cashless Arrangements; Data
Warehousing; Tax Audit tools and processed

%

103

Tax-for-Service/Improvement of Tax Harmonization
Implementation in Selected LGAs

%

104

Reviewing and Updating LGA Accounting and Audit
Manuals

%

105

Tax Harmonization Improvement Project in up to 5
pilot LGAs

%

106

Improved Tax Complaints Process Project

%

107

Lagos Informal Sector Business Management
Development Program

%

108

Enterprise Support Units in selected businessoriented LGAs

%

109

Technical assistance for the establishment of an
Investment Promotion Agency (IPA)

%

110

Support to the Ministry of Women Affairs, Lagos
State on the development of Women’s Economic
Empowerment (WEE) Strategy/Policy

%

111

Development of Industrial and Competitiveness
Policy

%

112

Partnership with Manufacturers Association of
Nigeria in commissioning a Study on EU (Economic
Partnership Agreement-EPA) and ECOWAS
(Common External Tariffs) Protocols and their impact
on the manufacturing sector in Lagos State

%

%

%

%

Business
support
services

64

65


Annexes

Support
to BMOs
and related
advocacy
and industry
coordination
GEMS 4

1

Skills
development
and capacity
building

Establishment
of new or
improved
products,
services or
processes

Business
enabling
environment
reform

1

4

1

Business
support
services

2

113

Process upgrading interventions

%

114

Product upgrading interventions

%

115

Functional upgrading interventions

%

116

Channel upgrading interventions

%

117

Supply chain management interventions

%

118

Mobile money interventions

%

119

Association strengthening

120

Policy and advocacy interventions

121

Skills development interventions
TOTAL GEMS

%
%
%
22

11

24

60

4

GEMS M&E Handbook

More Related Content

What's hot (19)

Similar to GEMS M&E Handbook (20)

More from Market Development Training (20)

Recently uploaded (20)

GEMS M&E Handbook