Proxy indicators Definition
This brief defines and explores the role of indicators as an integral part of a results-based accountability system. The brief shows how indicators enable decision makers to assess progress toward the achievement of intended outputs, outcomes, goals, and objectives.
About This Series
These short reports are designed to frame and contribute to the public debate on evaluation, accountability, and organizational learning.
I. Indicators Defined
An indicator provides evidence that a certain condition exists or certain results have or have not been achieved (Brizius & Campbell, p.A-15). Indicators enable decision-makers to assess progress towards the achievement of intended outputs, outcomes, goals, and objectives. As such, indicators are an integral part of a results-based accountability system.
II. Types of Indicators
Indicators can measure inputs, process, outputs, and outcomes. Input indicators measure resources, both human and financial, devoted to a particular program or intervention (i.e., number of case workers). Input indicators can also include measures of characteristics of target populations (i.e., number of clients eligible for a program). Process indicators measure ways in which program services and goods are provided (i.e., error rates). Output indicators measure the quantity of goods and services produced and the efficiency of production (i.e., number of people served, speed of response to reports of abuse). These indicators can be identified for programs, sub-programs, agencies, and multi-unit/agency initiatives.
Outcome indicators measure the broader results achieved through the provision of goods and services. These indicators can exist at various levels: population, agency, and program. Population-level indicators measure changes in the condition or well-being of children, families, or communities (i.e., teen pregnancy rate, infant mortality rate). Changes in population level indicators are often long-term results of the efforts of a number of different programs, agencies, and initiatives. In some cases, rather than providing information about the results achieved by interventions, population-level indicators may provide information about the context in or assumptions under which these interventions operate. For example, the overall level of unemployment provides important contextual information for job placement programs. In this case, monitoring the unemployment rate allows stakeholders to correctly interpret program results. Agency-level indicators measure results for which an agency is responsible; program-level indicators measure the results for which a program or sub-program is responsible. Agency- and program-level outcome indicators are often defined more narrowly those pertaining to the population as a whole; for example, they may measure pregnancy rates among teenage girls in a given county or among girls receiving a given set of services. Identification of appropriate indicator levels ensures that expectations are not set unrealistically high.
III. Criteria for Selecting Indicators
Choosing the most appropriate indicators can be difficult. Development of a successful accountability system requires that several people be involved in identifying indicators, including those who will collect the data, those who will use the data, and those who have the technical expertise to understand the strengths and limitations of specific measures. Some questions that may guide the selection of indicators are:Does this indicator enable one to know about the expected result or condition?
Indicators should, to the extent possible, provide the most direct evidence of the condition or result they are measuring. For example, if the desired result is a reduction in teen pregnancy, achievement would be best measured by an outcome indicator, such as the teen pregnancy rate. The number of teenage girls receiving pregnancy counseling services would not be an optimal measure for this result; however, it might well be a good output measure for monitoring the service delivery necessary to reduce pregnancy rates.