Definition of Benchmarking

A benchmark represents the best or ideal practice in a sector. Performance of organizations or programs may be compared with the benchmark in order to fix their current level of achievement and how much more they need to do in order to reach the desired level.

Usefulness of Benchmarks

• Benchmarks are measurable
• Benchmarks represent a standard which is fairly stable over a period and across a sector/ industry
• Benchmarks represent a realistic measure of excellence achieved in a field and not an imaginary state of affairs
• Benchmarks make comparison and planning for future performance easier
• Benchmarks make it possible to distinguish a less effective program from a more effective one because the same standard applies to both

Benchmarks may capture the following aspects of performance

• Regularity
• Reliability
• Economy/Cost Efficiency
• Quality

 Measurement of Impact

Impact of a program can be measured through the following.

• Comparison of the program group with control group (same time across space) (use of a true control group helps isolate the impact made by the treatment/ factor/ program, that is under study); here, the impact is known as the program effect or Yt - Yc (i.e., difference between outcomes of the treatment and control groups)
• Comparison of present situation with pre-program/ baseline situation (across time) (ex-ante v/s ex-post)
• Tracking of impacts over a long period (time series design)
• Comparison may be made (1) by conducting a number of survey rounds or (2) by recalling past data

Characteristics of Measurement Tools

What constitutes an ideal tool of impact measurement? In order to measure the impact accurately and reliably, the measurement tool must have the following characteristics.

• Simplicity: In an environment of participatory research, it is important the tool is simple enough to be understood and administered
• Sensitivity: Accurate measurement of impact depends on how sensitive is the instrument prepared for the purpose
• Causality: The instrument should also be able to detect what caused the impact
• Desirability: The instrument should be able to capture the intended and unintended impacts separately

Tools used for Impact Measurement

Impact Evaluation is usually conducted through a beneficiary survey. During the survey, information may be collected from the respondents in the form of coded responses to a structured questionnaire or in the form of personal data (e.g. age, sex, caste, weight, height, asset ownership, income, expenditure, etc).  Apart from the survey, the following tools can help the evaluator reach a conclusion about the impact made by a program:

• Physical observation of assets created during a program
• Narratives and qualitative inputs by the program beneficiaries or non-beneficiaries
• Secondary data collected from official records (files, registers, listings, etc)

 Analysis of the Impact

• Classification by degree of impact: high, moderate, low
• Causal analysis would bring out the causal relationship between any 2 sets of factors

Indicators of Impact

• Impact Indicators are an important part of the program log frame

Impact can be expressed in terms of percentages (e.g. proportion of total population having a particular characteristic) and in absolute numbers

Measuring the Impact of Social Accountability Efforts

The success of a social accountability mechanism, such as a social audit, depends on a host of factors such as: quality of the social auditors deployed, inclusiveness and simplicity of the audit process, etc. Impact evaluation can play a crucial role in identifying the success factors in the case of a particular social accountability exercise. Impact evaluation can also help quantify the impact of the exercise in terms of: (a) awareness it created in the community, (b) no. of cases of irregularities, corruption and mis-governance detected during the exercise, etc.

Comparison group designs can help understand the difference between two situations; one, where a social accountability tool was used to evaluate a program and another, in which traditional tools were used.