Information not my work

Signposts of Development

RBM in UNDP: Selecting Indicators

Contents

What indicators tell us about the wind

1. Introduction
2. Indicators Only Indicate
3. Types of Indicators
4. Qualitative and Quantitative Expressions of Indicators
5. Selecting Indicators
6. Indicators are Practical and Observable
7. A Shared Commitment With Partners
8. Baseline, Target and Timeframe
9. Indicator Data Collection
10. Conclusion

Figures
1. Assessing performance along results
2. School enrolment as an indicator of access to education

Tables
1. Examples of outcomes and outcome indicators
2. Indicators according to relationship between output, outcome and impact
3. Examples of output and outcome indicator scales
4. Qualitative observation of advocacy and upstream policy results
5. Selection criteria for indicators
6. Sample indicator monitoring plan

What indicators tell us about the wind

Who has seen the wind?
Neither you nor I.
But where the trees bow down their heads,
The wind is passing by.

– The Wind, Christina Rosetti, 1830-1894

If we want to know if the wind is blowing, the bowing treetops are a good indicator.

That the treetops are bending may tell us many different things: the coming of a change in seasons; that it is a good day for children to fly kites; or that bird nests may fall down. What it tells us depends on who we are and what we want to know.

For some, however, the fact that treetops are bending may not represent the most important thing they want to know about the wind. In order to steer his boat, the fisherman may need to know the direction of the wind. The farmer, on the other hand, may want to know how strong the wind is, because her crops might be ruined. Or perhaps the farmer is less interested in the wind than in the type of clouds, which may be a better indicator of coming rain.

Whatever the treetops tell us, they will never tell us why the wind is blowing. They will not explain the complex meteorological changes occurring in the atmosphere or reveal that the wind is coming from a storm far out at sea.

Indicators then, can tell us that a change we are interested in is happening. And indicators can be framed in a way that is most relevant to us. But they cannot explain why and how that change occurs. They can tell us the wind is blowing, but not why, to what effect or what we should do about it.

1. Introduction

Demonstrating progress towards results

Indicators are signposts of change along the path to development. Indicators are what we observe in order to verify whether – or to what extent – it is true that progress is being made towards our goals, which define what we want to achieve.

Indicators make it possible to demonstrate results. Indicators can also help in producing results by providing a reference point for monitoring, decision-making, stakeholder consultations and evaluation. In particular, indicators can help to:

• Measure progress and achievements; • Clarify consistency between activities, outputs, outcomes and goals; • Ensure legitimacy and accountability to all stakeholders by demonstrating progress; • Assess project and staff performance.

By verifying change, indicators help us demonstrate progress when things go right and provide early warning signals when things go wrong. This assists in identifying changes that need to be made in organizational strategy and practice. The continuous monitoring of indicators also facilitates effective evaluation.

When to use indicators

The purpose of indicators is to support effectiveness throughout the processes of planning, implementation, monitoring, reporting and evaluation – in other words, throughout the full spectrum of results-based management.

Indicators may be used at any point along the results chain of inputs, activities, outputs, outcomes and impacts. They may relate to the actual achievement of the result (target), to the current situation the partners are trying to change (baseline) or to progress or process (annual targets, intermediary benchmarks). Country offices (CO) may apply indicators to any other process or product that is useful to measure (e.g., the formation of new partnerships, the time spent on soft assistance initiatives or the delivery of activities in project work plans).

2. Indicators Only Indicate

Limits and risks

Indicators only indicate; they do not explain. Determining that change has occurred does not tell the story of why it has occurred. Indicators constitute only one part of the logical and substantive analysis needed for development efforts to succeed. In addition, success requires common sense, sound managerial judgment, leadership and creativity – none of which can be replaced by the use of indicators.

The use of indicators can be made into an elaborate science demanding a major workload. Using a large number of different indicators, however, has no merit in itself. The key to good indicators is credibility – not volume of data or precision in measurement. A quantitative observation is no more inherently objective than a qualitative observation. Large volumes of data can confuse rather than bring focus. It is more helpful to have approximate answers to a few important questions than to have exact answers to many unimportant questions.

Practical proxy for change

The fundamental challenge with indicators is to meaningfully capture key changes or “results”. This is accomplished by combining what is substantively valid with what is practically possible to monitor.

In development, it is often difficult to make objective and exact observations of the complex development changes we are addressing. Instead, we frequently rely on observations that approximate intended changes. We use indicators that are commonly understood to be closely related, e.g. share of social expenditures in a government budget as “proxy” for poverty orientation of national policies, or proportion of parliamentarians who are female as “proxy” for empowerment of women in national decision-making processes.

3. Types of Indicators

Three results indicators

Different types of indicators are required to assess progress towards results. Within the RBM framework, UNDP uses at least three types of indicators, also known as results indicators:

• Situational (impact) indicators, which provide a broad picture of whether the developmental changes that matter to UNDP are actually occurring (impact indicators and situational indicators are essentially the same, although the former may be more specific and the latter may be more generic); • Outcome indicators, which assess progress against specified outcomes; • Output indicators, which assess progress against specific operational activities.

Figure 1 illustrates the linkages between situational, output and outcome indicators for assessing performance results in terms of results. It also highlights the level of management – project, programme or senior country office – that would find a particular type of indicator most useful, allowing for a degree of overlap among the levels and types of indicators.
Figure 1. Assessing performance along results

Situational indicators

Situational indicators describe the national development situation. They relate to the Millennium Development Goals and the SRF Goals and Sub-goals, and reflect long-term development results, or impact. Situational indicators provide a broad picture of country development status (macro baseline). They are most useful to the country office senior management, informing the level at which senior management interacts with partners and develops strategies.

To find examples of situational indicators, refer to the National Human Development Report (NHDR), the Common Country Assessment (CCA) and national statistics. Specific examples of situational indicators include the signature UNDP-initiated development indicators such as the human development index (HDI) and the human poverty index (HPI) as well as others developed by the OECD and adopted by the United Nations system.

Outcome indicators

Outcome indicators help the organization and country offices think strategically about the key results or outcomes they want to achieve. They help verify that the intended positive change in the development situation has actually taken place. Outcome indicators are designed within the SRF framework and the Country Programme, and they are most useful to the country office programme managers who liase both with the CO senior management on progress towards outcomes and with the project-level management on the contribution of outputs to outcomes.

• An outcome indicator for an advocacy activity aimed at policy change in governance institutions may include observing parliamentary passage of a desirable legal change, and also observing if the new law is backed up by an allocation of financial resources by a certain date.

Examples of outcome and outcome indicators adapted from the Strategic Results Framework (SRF) are listed in Table 1.

|Table 1: Examples of outcomes and outcome indicators |
|Expected outcome |Outcome indicator |
|The policy, legal and regulatory framework reformed to |Estimated number and proportion of the population with access to the |
|substantially expand connectivity to information and communication|Internet, disaggregated by gender |
|technologies (ICT) | |
|(Poverty, G2-SGN2-SASN3) | |
|Increased effectiveness of parliament to perform its oversight |Mechanisms for accountability such as a Public Accounts Committee are|
|functions |in place to ensure legislative oversight |
|(Governance, G1-SGN2-SASN1) | |
|Fair and efficient administration of justice |Level of public confidence in the justice system |
|(Governance, G1-SGN2-SASN3) | |
|The national policy framework reformed to achieve universal access|Proportion of the poor covered by institutionalized social security |
|to basic services |and insurance schemes (pensions, health, unemployment) or transitory |
|(Poverty, G2-SGN2-SASN2) |safety nets (programmes for temporary unemployment, food-based or |
| |cash transfers) |
|Women’s organizations enabled to advocate and monitor improvements|Existence of national legislation and measures to provide an enabling|
|in the status of women |environment for women’s non-governmental organizations (NGOs) |
|(Gender, G4-SGN1-SASN3) | |
|Systematic monitoring of progress in linking national development |A comprehensive monitoring and evaluation system created and applied |
|targets with globally agreed goals |to track progress in the country’s human development in the context |
|(UN Support, G6-SGN1-SASN2) |of PRSP |

Source: Adapted from the Strategic Results Frameworks of Gabon, Kyrgyzstan, Nepal, Tanzania and Venezuela, 2001.

Output indicators

Output indicators help to measure and verify the production of outputs. Outputs are tangible results that can be delivered within a short timeframe. This means that the output itself may be measurable and may clearly indicate how to verify that it has been produced. Output indicators are most useful to project managers, who are responsible for the production of outputs and their relevance to the outcome in question.

• For the output “district school teachers trained”, for example, an output indicator could be created simply by adding a target such as “50 teachers trained by end 2001”.

• In the SRF, outputs have annual targets. For the outcome “increased public debate on sustainable human development (SHD)”, one UNDP country office used the following outputs: “seminars for Ministers of Parliament, journalists […and others]”, “specialized programmes in the media” and “two National Human Development Reports (NHDRs)”. For “Specialized programmes in the media” the target is “promotion of SHD in at last 4 media programmes and Resident Representative interventions on the media, in one year”.

Table 2 illustrates examples of how indicators may be applied for output, outcome and impact for various types of programmes adapted from the SRF.

|Table 2: Indicators according to relationship between output, outcome and impact |
| Area |Water supply |Women’s empowerment |Environment |Human rights |
|Result | | | | |
|Output |Number and type of wells |Number of loans given and |Number of species planted |Number and category of people |
| |installed |repaid as agreed |properly and surviving |given training or other types of |
| | | | |support |
|Outcome(s) |The number and proportion |Percentage of women with |New areas reforested and |More active censure of |
| |of population with |increased disposable income, |sustainable agricultural |politicians and law-enforcing |
| |sustained availability of |expanding their options |practices applied |agencies |
| |clean water for proper |towards diverse social and | | |
| |domestic use |economic roles |Better economic |Greater financial allocation by |
| | | |opportunities for |government to monitor and address|
| | | |indigenous or isolated |human rights abuse |
| | | |communities | |
|Impact |Reduction in ill health and|Improved economic control, |Retention or increase in |More transparent, accountable |
| |mortality |choice and status with |forest areas |state behaviour with reduction in|
| | |respect to men | |political arrests |

4. Qualitative and Quantitative Expressions of Indicators

Signals and scales

Indicators can comprise a variety of types of “signals”—in other words, how the indicator is expressed—such as numbers, ranking systems or changes in the level of user approval. A signal also features a “scale” of observation. For example, the indicator “65 per cent of enrolled students graduate secondary school” features a percentage signal with a scale of 65 per cent.

Signals and scales lend themselves to indicators that express qualitative and/or quantitative information. Quantitative indicators are numerical. Qualitative indicators use categories of classification. (Some analysts define qualitative indicators as those that are based on individual perceptions, e.g. as given in response to survey questionnaires).

|3. Examples of output and outcome indicator signals and scales |
|Qualitative/Categorical Indicators |
|Signal |Scale |Outcome indicator |Output indicator |
|Existence |(yes/no) |Local governance act passed/not |Policy recommendation submitted/not |
| | |passed |submitted |
|Category |(e.g. x or y or z) |Level of SHD policy focus “high”, |Poverty analyzed in “region east”, |
| | |“medium” or “low” |“west” or “nationally” |
| |
|Quantitative/Numerical Indicators |
| |Scale |Outcome indicator |Output indicator |
|Number |(e.g. 1, 20 or 5,000) |Number of new jobs created in small |Number of entrepreneurs trained |
| | |enterprise sector | |
|Percentage |(e.g. 12% or 95%) |Percentage share of rural population |Percentage share of government budget |
| | |with access to basic health care |devoted to social sectors |
|Ratio |(e.g. 1/3 or 125 per 100,000) |Ratio of female to male school |Ratio of trained female to male members|
| | |enrolment |of parliament |

No one type of indicator or observation is inherently better than another; its suitability depends on how it relates to the result it intends to describe.

Qualitative measurements of change

When a result is qualitative, it is still possible to develop an indicator that offers some measure of the magnitude of change. For example, if the proportion of people who perceive of local government management as “very participatory” increases from 40 per cent to 65 per cent over a certain period of time, this increase provides some measure of the degree of qualitative change.

This kind of numerical expression of qualitative considerations may also be obtained through indicators that use rating systems that rank, order or score given categories of attributes. An example might be an average, as follows: “With regard to responsiveness to their own needs, on a scale from 1 to 10, people in rural areas give an average score of 2.5 to central government and 6.2 to their local council”. Another alternative is to present a result as a distribution, as follows: “With regard to responsiveness to their own needs, the proportion of people in rural areas who rate central government 3 or less is 60 per cent, and the proportion who rate their local council 5 or better is 65 per cent”. A special variant of a distribution is captured by the Gini Coefficient, which measures dispersion on a scale from 0 to 1. Where all observations are the same, the Gini coefficient is 0; where they are all different the Gini is 1.

In the example above, outcome-level change is measured in response to public action. This is done through a “proxy” measure: changing levels of end-user approval (or client satisfaction). Such an indicator is particularly helpful when the public actions involve capacity development or direct public service delivery. The perceptions of end-users regarding public service delivery gets straight to the issue of whether the services are wanted, useful and effectively delivered. The satisfaction of end-users (or clients) has the advantage of some comparability. Results may be compared:

• Between different kinds of service; • Between separate locations; • Over time.

This approach is not without its problems, however. Clients may not always be easy to identify, and their perceptions of satisfaction with services is subject to influences other than the service itself.

Combining qualitative and quantitative

Indicators may combine quantitative and qualitative observations. In a programme aimed at increasing access to social services, for example, the indicator includes a baseline in which 10 per cent of people had access to a particular kind of service (now or some time in the past). The target is for 30 per cent to have access to an improved service, by a certain date in the future. In another example, an advocacy activity aimed at policy change may use an indicator that includes observing parliamentary passage of a desirable legal change (yes/no) and that also includes the new law being supported by an increased allocation of 15 per cent more resources than was previously the case, by a certain date.

An index is a composite indicator, formed by amalgamating two or more different measures into one, such as the Human Development Index. With indices, “weight” must be attributed to components according to their relative importance. In other words, if one of the component attributes is inherently more important than another, it should be assigned a heavier weight or share of the combined measure.

An indicator may be expressed as something to be minimized or something to be maximized, e.g. working to maximize the number of doctors per population of 1,000 is the same as working to minimize the number of people per doctor.

Advocacy and policy results

Qualitative assessments are often required when measuring the success of UNDP efforts related to advocacy, coordination and upstream policy advice and dialogue.

|Table 4. Qualitative observation of advocacy and upstream policy results |
|Policy result type |What can concretely be observed, Data sources |
|Change in policy priorities |Change in relative shares/composition of budget |
| |Cabinet decision, ministerial policy declarations |
| |Accession/signature of UN conventions and international accords |
|Enactment of new legislation |Passage by legislative body, confirmed for example by the Parliamentary Gazette or |
| |Hansard |
| |Record of Cabinet decisions |
|Establishment of new institutions, practices |Establishment of posts |
|and programmes |Allocation of budget |
| |Commencement/availability of service |
| |Revision of public service code |
| |Conditions and prices of service delivery |
| |Client charter commitments |
| |Client satisfaction levels (as surveyed) |
|Access to information |Availability of publications, records in print, on Internet |
| |Access-to-information laws passed |
|Participation in decision-making |Number of civil society organizations (CSOs) invited to comment on policy drafts |
| |Discussion forums convened |

If indicators are multi-dimensional, they often need to be disaggregated, or broken down, in order to reveal differences between their various components. This separates findings related to gender, geographic location, socio-economic group (age, ethnicity, religion, income level). The level of disaggregation should reflect the area of operation or target group pertaining to a goal or set of interventions.
5. Selecting Indicators

The impact of selection

What type of indicator is best to use? The choice makes a difference. If the wrong thing is measured, or if it is measured in the wrong way, the data may be misleading and the quality of decisions could be affected. The choice also may have unforeseen consequences—some positive some negative. When a country changed its system for hospital funding from one based on the number of drugs doses administered to one based on the number of patients treated, doctors became more friendly to patients but also wrote fewer prescriptions for medicine. And when a police force changed its result indicator from number of arrests to number of convictions, for example, constables became less inclined to detain people on the basis of vague suspicion.

The challenge in selecting indicators is to find measures that can meaningfully capture key changes, combining what is substantively relevant as a reflection of the desired result with what is practically realistic in terms of actually collecting and managing data.

The SMART way to select indicators

The following criteria and questions may be helpful in selecting indicators. As a memory aid, the acronym “SMART” summarizes key criteria, asking “Is the indicator specific, measurable, attainable, relevant and trackable?”

Specific: • Is it clear exactly what is being measured? Has the appropriate level of disaggregation been specified? • Does the indicator capture the essence of the desired result? • Does it capture differences across areas and categories of people? • Is the indicator specific enough to measure progress towards the result? For example, using the indicator “increase by 20 per cent in number of criminal complaints filed” may reflect a more effective justice system OR an increase in crime.

Measurable: • Are changes objectively verifiable? • Will the indicator show desirable change? • Is it a reliable and clear measure of results? • Is it sensitive to changes in policies and programmes? • Do stakeholders agree on exactly what to measure?

Attainable: • What changes are anticipated as a result of the assistance? • Are the result(s) realistic? For this, a credible link between outputs, contributions of partnerships and outcome is indispensable.

Relevant: • Does the indicator capture the essence of the desired result? • Is it relevant to the intended outputs and outcome? To judge the relevance of indicators, the CO may have to identify the target groups and their needs, expectations and criteria. • Is the indicator plausibly associated with the sphere of activity?

Trackable: • Are data actually available at reasonable cost and effort? • Are data sources known? CO should establish realistic principles, mechanisms and responsibilities for data collection. • Does an indicator monitoring plan exist?

Be sensible and practical in applying these criteria. No one indicator will satisfy all criteria equally well. Ultimately, the choice of indicator is determined through a holistic assessment of validity and practicality. The selection of indicators is an iterative process, building on consultations between programme managers, stakeholders and partners. The process of selecting an indicator takes several steps including brainstorming ideas, assessing each one and narrowing the list (using the criteria above) and, finally, making an indicator monitoring plan.

|Table 5. Selection criteria for indicators |
|Indicator selection |Outcome |Poor proposal for an indicator |Why indicator is inadequate |Possible refinement of indicator |
|criteria | | | |(within a given timeframe) |
|Specific |Better understanding of|Government officials, social |Who one should consider a social |Number of parliamentary or media |
|or “Precise meaning” |UN mandates and UN |leaders speak about UN. |leader is arguable. No |references to (specified) UN |
| |work. | |baseline/target. |conferences or resolutions up from 10 |
| | | | |to 30 per year. |
|Specific |Job creation through |Micro-capital finance available|Availability of finance is a means, |Increase from 200 to 500 in number of |
|or “Valid “ |micro-enterprise. |in 5 regions, up from 2. |not an end result. The purpose is to |people employed by trained |
| | | |create employment growth. |micro-enterprises. |
|Measurable |Enhanced capacity of |Improved job prospects for |Job prospects can only be assessed |Increase in school enrolment rate from |
|or “Practical” |school planning system.|those who leave school early. |when students graduate – many years |85% to 95%. |
| | | |from now. No baseline/target. | |
|Attainable |Transparency in public |Reduced number of corruption |Transparency awareness may (at least |Policy and practice changed to make |
|or “Clear direction” |sector finances. |cases. |initially) lead to number of |protocols of tender board meetings |
| | | |prosecutions going up – not down. No |available for public inspection |
| | | |baseline/target. |(yes/no). |
|Relevant |Local development |Increase from 50 to 200 in |Beneficiaries do not care about how |Percent of local development funds |
|or “Owned” |planning responds to |number of community funding |many proposals are received, but how |actually allocated to community |
| |priorities of the poor.|proposals submitted to local |many are approved. |initiatives (submitted by NGOs, CSOs) |
| | |planning authority. | |increased from 25% to 50%. |
|Trackable |Professional standards |Quality of journalistic |Too many elements in the indicator, |Increase in number of media |
|or “Data availability” |and independence of |coverage in terms of |all open to subjective judgment. No |independently established and financed |
| |media strengthened. |independence, ethics, |national data collection. No |(from 2 to at least 6). |
| | |professional standards as well |baseline. Can have professional |Public survey results with satisfaction|
| | |as coverage of vulnerable |standards without covering vulnerable|with ethical media coverage of at least|
| | |groups. |groups. |40%. |

6. Indicators are Practical and Observable

A practical process

The process of selecting indicators can help identify the core issues in a UNDP intervention and translate intangible concepts into more concrete and observable elements. This process may assist managers and stakeholders in carrying out the initial problem analysis and in articulating results expectations.

Even a carefully selected, clearly defined indicator is of little use unless it is actually put to use. A critical test of an indicator is how practical it is to monitor. Thinking about an indicator is one thing; actually finding, recording and presenting the data is another. Indicators need to be approached as a practical tool, not merely as a conceptual or aspirational exercise.

From an RBM perspective, one of the main purposes of indicators is to validate results based on objective observation, facts or actual experience. This is far more concrete than relying on the perceptions of individuals responsible for implementing programmes. In this way, the use of indicators lends credibility to UNDP’s results reporting under Strategic Results Framework/ Results-Oriented Annual Report (SRF/ROAR).

Observational criteria

Along with a commitment to actually monitor and use indicators, their effectiveness requires a clear idea about what will be observed and how to do so. This underlies the observational criteria associated with an indicator.

It is important to identify the essential observational criteria, as in the following examples:

• Institution X is operating more effectively/efficiently; • SHD objectives are better integrated/included/reflected in policies and plans; • Measurable increase in client satisfaction with ministry Y’s services.

For these statements to be valid as indicators, there must be a definition of how one will measure effectiveness, integration or satisfaction – and how data are to be identified in actual practice. An indicator lacking clarity about what and where to observe is meaningless. We need to specify exactly what should count and how it should be counted, with the exception of existing standards that are commonly known. In terms of the SRF/ROAR, it is critical that country offices select indicators for which they know data will be available. Table 6 provides further illustration of how concrete indicators can be crafted with more clarity within an indicator monitoring plan.

|Table 6. Sample indicator monitoring plan |
|Intended result |Result indicator |Data source |Frequency of |Monitoring responsibility |
| | | |observation | |
|Outcomes: | | | | |
|Enabling environment for |Increase in number of CSOs participating |Registrations for |Annual |Coordination Council for |
|participation by the poor in |in national development strategy |Annual Poverty Forum| |NGOs |
|decision-making |formulation process (by year) | | | |
| | | | | |
| |New staff taken on by small-scale | | | |
|Community prospects improved |enterprises (SSEs) trained (number, by |Questionnaire sent |Half-yearly |National Chamber of |
|through income generation |date) under UNDP programme |trainees 6 months | |Commerce |
| | |after training | | |
|Outputs: | | | | |
|Civil society organizations’ (CSOs)|NGO staff completing training course in |Project records |Quarterly |Project Director |
|analytical capacities strengthened |poverty analysis (number, by date) | | | |
| | | | | |
|Training facilities available to |SSE advisory centres functional (able to | | | |
|SSEs |offer training) in 3 regions (by date) |Project records |Continuous |UNDP National Programme |
| | | | |Officer (NPO) |

7. A Shared Commitment With Partners

External data sources

Many agencies, managers and programmes contribute to the collection of data through the monitoring of indicators. Concerted action by these various groups and individuals is required to ascertain that change has occurred. By sharing data collection with these “external data sources”, UNDP is not required to undertake data collection by itself. Reliance on external data sources is becoming more important with the increasing emphasis on results-based management. Indeed, the individual contributions of partners can rarely, if ever, be objectively determined, meaning that successes are necessarily shared.

When UNDP contributes towards outcomes (development results) in partnership with others, not all monitoring responsibilities fall to UNDP. All partners must agree on which indicators to monitor and who among the partners will take responsibility for monitoring those indicators.

Sharing builds partnerships

The sharing of indicators can be a valuable exercise in partnership. A commonly agreed upon set of indicators reflects a shared understanding of problems, goals and strategies. (Such sharing of indicators is itself an indicator of partnership.) As mentioned above, the quest for finding common indicators is part of forging consensus among partners. An outsider looking at a set of indicators may ask: who decided that these are the most important ones? Sharing ensures greater agreement among all partners. Without such cooperation, UNDP might use one set of indicators, national authorities a different set and an external partner yet another – reducing the likelihood that all parties are pulling in the same direction.

8. Baseline, Target and Timeframe

Making indicators meaningful

Indicators require a baseline, target and timeframe in order to be useful in verifying the results of a development intervention. This makes it possible to demonstrate change over time. The baseline is the situation before a programme or activity, and is the starting point for results monitoring. The target is the situation expected at the end of a programme or activity. Between the baseline and the target there may be several milestones that correspond to expected performance at periodic intervals. The timeframe refers to observations taken at specified points in time or within a given period of time.

Figure 2: School enrolment as an indicator of access to education

In the figure above, wider access to education is our intended result. The percentage of school enrolment may provide a good outcome indicator. To monitor results, we may start with a baseline of 65 per cent enrolment in 2002 and a target of 80 per cent enrolment in 2005. Along the way, we may establish that 70 per cent is a good result for end of 2003 and 75 per cent for end of 2004.

Establishing a baseline

Ideally, the baseline should be gathered and agreed upon by stakeholders when a programme is being formulated. For many ongoing activities, however, no baseline was specified at that time. In some cases, it may be possible to retroactively ascertain approximately where one was when the programme started, perhaps from data included in past annual review exercises.

When retroactive sources of data do not exist, it still may be possible to obtain a measure of change over time. For example, to establish a baseline pertaining to local governance one might ask a number of people: “Compared to three years ago, do you feel more or less involved in local decision-making?” A clear tendency among respondents towards “more” or towards “less” provides a valid indication of whether change has occurred or not.

When it is impossible to retroactively establish any sense of change, establish a measure of where one is now. This will at least allow for the assessment of change in the future.

Establishing targets

The key to establishing targets is realism. Target-setting must be based on a thorough review of the factors that influence the development problem being addressed, what partners are doing, and what degree of change can realistically be associated with the UNDP contribution. Factors to consider in establishing targets include:

• Past trends, i.e., change observed over previous periods; • How well others have done; • Limits to progress; and • The existence of objective international, sectoral or other quality standards.

Targets need to be commensurate with UNDP’s sphere of influence, resources, expertise and partnerships. Overly ambitious targets lead to disappointment when they are not met.

9. Indicator Data Collection

Practical considerations

Indicators should be selected because they are relevant to the results being monitored, and not simply because they are easy to track. This ensures that indicators provide data that are appropriate and useful.

Counterpart institutions receiving UNDP support for capacity development often carry out the collection of data. When entering into a programme of support, it is important to specify exactly how this work will be completed and who will carry it out. Ultimately, UNDP and other partner institutions have a responsibility to ensure that this work gets done.

How to obtain data

Sometimes a simple instrument introduced within the programme management process can provide useful information. For example, to gather data on the indicators “new firms started yes/no” or “purchase of raw materials increased/decreased by X amount”, one could send a questionnaire to entrepreneurs six months after they have completed training under a small business development programme. Depending upon what is asked, the responses could reveal whether or not entrepreneurs had started new firms, taken on new staff or how much they had invested in local raw materials and supplies.

Information to track some indicators is available from public records or common knowledge among good managers, such as the passage of a critical piece of legislation. When looking at outputs (assessing progress against specific operational activities rather than overall development results), information is often available from internal records that detail the number of people trained, policy recommendations submitted to national authorities or policy seminars convened, for example.

In other cases, information is not immediately available and may require the development of instruments and/or capacities to capture information.

Sources and tools for data collection include, among many others: • Administrative records of client service; • Statistical surveys; • Awareness/attitude questionnaires; • Expert panels, trained observers; • Focus groups and key informant interviews.

Sometimes UNDP obtains data from internal records—for example, number of people trained, policy recommendations submitted to national authorities or seminars held. Often, UNDP obtains data directly from external partners. This kind of data collection is generally commissioned, not carried out, by UNDP. When commissioning data collection by partners, UNDP’s major concern is quality assurance, which might be affected by the selection of participants in expert panels or the choice of sampling methods used in questionnaires.

10. Conclusion

Indicators are signposts of change. We should bear in mind, however, that indicators are only intended to indicate, and not to provide scientific “proof” or detailed explanations about change. In addition, we should avoid the temptation to transform the measurement of change into a major exercise with a burdensome workload. It is development change we seek to influence that must remain the driver—not the indicator. Measuring change should not take precedence over programme activities that generate the changes to be measured.

Part of the toolbox

Indicators enable us to verify the status of outcomes, or development changes, that UNDP seeks to influence as well as the progress of products and services for which managers are responsible. The purpose of indicators is to support effective programme planning, management and reporting. Indicators not only make it possible to demonstrate results, but they also can help produce results by providing a reference point for monitoring, decision-making, stakeholder consultations and evaluation. The use of indicators is integral to good management. Indicators are not merely relevant to scientists, statisticians or staff at UNDP headquarters. Indicators support the entire process of managing for results at every level of the organization. They help all of us “keep our eyes on the ball”.

Indicators are part of the toolbox that managers and staff need to bring to their work. This is especially true now that UNDP and other development agencies are being called upon to demonstrate effective results.

Limitations

The critical issue in selecting good indicators is credibility, not precision in measurement. Indicators do not provide scientific “proof” or detailed explanations about change. There is no objective truth or certainty to information represented through indicators. But indicators that are carefully considered and shared among partners are much better than guesswork or individual opinion. An indicator that provides relevant data about progress towards results is very useful. At the end of the day, it is better to have approximate information about important issues than to have exact information about what may turn out to be trivial.

Applications

The important thing, in the end, is how indicators are used as part of the decision-making process. Indicators are intended to provide data that will help managers and staff make better decisions, achieve results and improve organizational effectiveness.
Acronyms

CCA Common Country Assessment
CO Country Office
CSO Civil Society Organization
NHDR National Human Development Report
NPO National Programme Officer
NGO Non-Governmental Organization
RBM Results-Based Management
ROAR Results-Oriented Annual Report
SRF Strategic Results Framework
SHD Sustainable Human Development
SSE Small-Scale Enterprises
UNDAF United Nations Development Assistance Framework

Glossary

Indicator: Signal that reveals progress (or lack thereof) towards objectives; means of measuring what actually happens against what has been planned in terms of quantity, quality and timeliness. It is a quantitative or qualitative variable that provides a simple and reliable basis for assessing achievement, change or performance.

Results-Based management (RBM): A management strategy or approach by which an organization ensures that its processes, products and services contribute to the achievement of clearly stated results. Results-based management provides a coherent framework for strategic planning and management by improving learning and accountability. It is also a broad management strategy aimed at achieving important changes in the way agencies operate, with improving performance and achieving results as the central orientation, by defining realistic expected results, monitoring progress toward the achievement of expected results, integrating lessons learned into management decisions and reporting on performance.

Results-Oriented Annual Report (ROAR): The principal instrument for reporting on performance and progress of results achieved on the entire range of UNDP interventions by operational units. In aggregate, the ROAR provides UNDP with a comprehensive assessment of key results achieved and a review of the resources at the organizational level. It is intended to provide a basis for feedback and continuous adjustment.

Strategic Results Framework (SRF): Represents the development hypothesis including those results necessary to achieve a strategic objective and their causal relationships and underlying assumptions. Establishes a basis for measuring, analyzing and reporting results of the operating unit. Can also be the overall aims and objectives of a country’s approach to development based on analysis of problems, and including a statement of priorities. For UNDP, the document that describes the results for an operating unit in terms of outcomes, outputs, partnerships and indicators with specified Goals, Sub-Goals and Strategic Areas of Support.
-----------------------

Reduction or Removal of Problem

SUSTAINED IMPACT

OUTCOMES

RESULTS

RESULTS

Output Indicators (effort)

Outcome Indicators (effectiveness)

Long-term Impact Indicators

Situational
Indicators

DEVELOPMENT PROBLEM / ISSUE DEVELOPMENT PROBLEM / ISSUE VELOPMENT PROBLEM / ISSUE

INPUTS

ACTIVITIES

OUTPUTS

M & E

Senior Management Level
(CCA/UNDAF)

Project Management Level

Programme Management Level

Senior Management Level
(CCA/UNDAF)

Primary Level Concerned

BASELINE TARGET (When we begin) (When we finish)

1 2 3 4 5 6 7 8 9 10

Very Low Low Medium High Very High

Alternative Indicator
SCALES

0% 10% 20% 30% 40% 50% 60% 70% 80%

Year 0 Year 1 Year 2 Year 3

2001 2002 2003 2004 2005 2006

TIMEFRAME

(Categories)

(Rating/score)

(Percentage)