Select the most important measures

Now you need to decide which three to five measures are the most important for your program.  You could do this by looking at all of the measures on your list for How much did we do?  How well did we do it?  Is anyone better off?

1. Identify the measures that will regularly have good data available.
2. Of the group of measures with good data, decide in order which ones you first choose to talk about your program.
   Ideas that might help you decide which measures are more important than others:
   • Which measures are easy to explain and would be understood by a wide variety of individuals?  For example, which measures would you use if you had five minutes with a stranger to describe your program and what makes it important?  These measures have communication power
   • Measures often look like at different parts of the same concept.  For example health care quality could be assessed using: (a) time with a health care provider, (b) percentage of recommended services that are provided, (c) using a health care patient's primary language, or (d) responding to patient questions.  If you have a group of measures, you might look for the one that could be used to represent the entire group.  You might also look for a measure that responds to your program the same way as the others do.  These representative measures have proxy power.
3. When you have organized your list of measures with good data in order of importance, your primary measures are the top three to five on the list; all of the others are your secondary measures.

Other thoughts on performance measures:

• Usually, programs are designed to help a group of people that is bigger than the number of customers it can serve.  The Results Accountability method seperately addresses the well-being of the larger group and the well-being of a program's customers.  Population accountability uses indicators to assess the status of a target population.  If a program serves a large proportion of the people that it is intended to help, you might be able to use a performance measure as an indicator.  This double purpose is helpful, because you will probably have better data for performance measures than for your indicators.
• You may identify a performance measure that wouuld be really excellent, except that there is no data for it.  You might do a couple things if this happens.  One would be to come up with a creative way of obtaining the data.  You might get some ideas if you talk with program managers, executives, customers, family members, program and/or community leaders, or state support staff.  The other strategy would be to start a data development agenda.  The purpose of this agenda is to help you keep track of the data you would like to have and how you might be able to obtain them.  You could use such an agenda during program design meetings, for instance, to find ways to fill important gaps in your data.

How are you doing on the most important of these measures?

Your performance measures will be more useful if you compare your results to other similar information.  You might consider putting performance measures for one period next to:

• past performance within the same activity, program or agency
• performance by other similar activities, programs or agencies, and/or
• standards for this activity or program, if they exist.

Your results might not be positive.  Accountability is as important when you do not show progress as when you do; it might be more important.  These reports are hard to make, though, because of the possible consequences.  You and program leaders can support accurate performance measurements by:

1. creating a healthy, non-judgmental organizational environment,
2. starting small, and
3. starting performance measurement from the bottom-up and top-down simultaneously.

Which of your partners have a role to play in doing better?

Now you need to ask people to help think about how to move your program forward.  This group should include the program managers and customers who have helped you work througuh the process so far.  You might also consider adding program staff, people from your customers' families, and community leaders.

This circle of advisors and advocates should include people inside and outside your organization who can help your program be more successful.

What works to do better, including no-cost and low-cost ideas?

Programs can always improve, and each use of your performance measures will give you information about where you might want to make program changes.  The question now is what would work better than what you are already doing?

Your partners may have ideas, and you might find more from other programs, from your own earlier successes or failures, or from state or federal agencies, suc as the Centers for Disease Control and Prevention (CDC).  As you develop a list of ideas, discuss these questions with your partners:

• What can be done that has no cost?  What could be done at minimal cost?
• Which partners can do which parts?
• What DOES NOT work?

You could also use these criteria to decide which ideas are more likely to be successful:

• consistence of values with your existing program and with the community
• detail clarity, such as who (customers and staff), whec, where and how
• support for activities or start new ones
• feasibility and affordability

If you need more information about a specific strategy, add this need to your research and information agenda.

What do you propose to do?

Your action plan should include:

• A list of the program changes you want to make this year, the changes you will make next year, and those that will happen in two to ten years.
• A description of how these chhanges fit into the existing program, and into the community system of services and agencies.
• Information about which partners will be responsible for which program components (new and existing).
• Budget information, including components that will not require more funds.
• Expected performance measures for new and existing program components.
• If you need data to fill an information gap (e.g., a data development agenda) and/or need more information about a specific program idea (e.g., an information and research agenda), list and develop budgets for these activities.