data that will support the business case; and readiness of the workforce and management teams to participate and support the proposed programs.

For management to make decisions in a timely manner, the data to be considered should be valid and reliable and may need to include projections based on data-driven assumptions (e.g., as related to projected return on investment) and should not be overwhelming. Moreover, there must be reasonable assurance to management that the data collected are representative, accurate, and reliable, thereby supporting their use for decision making that prompts action.

Measurement for Accountability

Measurement can also be used for accountability. One method of accountability for achievement of program objectives is periodic reporting of a set of measures created a priori. Moreover, program staff may assume accountability in a very proactive manner when the measures against which they are held accountable are known in advance. These measures may include process measures, but for the purposes of accountability, most of them will be outcomes or results-type measures. They are also useful for monitoring overall program performance. Measurement accountability should be reported openly, however, so that it can be used for performance comparison.

Assurance that the measures used for accountability are accurate and valid requires a focus on a few vital measures. In addition, the measurement process may need to include external staff or independent audits and be appropriately adjusted for validity. Furthermore, the creation of the measures themselves should be done in a collaborative manner so that agreement exists on the measures themselves.

Measurement for Improvement

The Plan-Do-Study-Act (PDSA) cycle is a good example of a measurement for improvement strategy (Langley et al., 1996). This approach includes data collection and measurement that identifies potential problems, barriers, or opportunities for improvement; facilitates the implementation of improvement initiatives; and follows with data collection to measure the improvement or change that has taken place. Following such an approach, measures and data used for improvement should be simple, easy to implement, and collected and reported in effective and efficient time frames. In addition, all data analyses should be capable of specific as well as centralized analyses.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement