This presentation provides an overview of the Government Performance Logic Model – the methodology utilized by The Performance Institute to promote performance management in government agencies.

Today I will be walking you through the logic model.  Please feel free to ask questions and seek clarification as we proceed. Before I jump to the heart of the presentation, I would like to provide you with more information about what The Performance Institute does, and why we are involved in the promotion of this methodology.  Should you need additional information, please feel free to contact us.  This slide provides our contact information.

The Institute is a private think tank dedicated to studying and improving the performance of government and non-profit organizations.  The Institute is involved in analyzing best management practices for dealing with emerging public policy issues.  Our focus has been to bring government leaders and managers together to explore common management challenges.  We review programs at the federal, state, and local levels of government, in addition to those of government partners, to seek out examples of dramatic change and improved performance.  Through this research, we develop best practice guides, how-to reports, training curricula, and consulting methodologies to promote effective management practices across the country.

In analyzing and sharing best practices in government, the Institute conducts four kinds of programs: 1) Conferences facilitate a peer-to-peer exchange of ideas, solutions, and lessons learned. 2) Our best practice research is condensed into applied training curricula for building the skills and capacities of government managers to implement best practices in their own programs.  The Institute provides in-house training on the logic model methodology.  3) The Institute conducts research and surveys on targeted questions and provides the results through published reports and e-newsletters with updates on new mandates and best practices. 4) Finally, many organizations seek out The Performance Institute’s help in a consulting capacity to design, implement, and evaluate change projects and management improvement initiatives. Now I’d like to move on to the topic of government performance, and discuss how things look today.

How do government managers and their staff members view performance-based government?  As the comments shown here indicate, they are pretty cynical. They complain that budget cuts are administered across the board, without regard to which programs are or are not producing the intended results.  SES performance plans are viewed as shams because they are in paragraph form and do not provide specific guidelines for the measurement of individual performance.  A-76 studies take too long.  By the time they are released, they are not useful.  Performance measurement systems contain too many indicators, burdening managers with extra work without being helpful.  Congress appears to be “out of touch,” and politicals are viewed negatively.  I understand these concerns.  In most cases, past approaches to performance-based government have missed the mark, leaving line managers frustrated with a broken system.  And they are the most important players in management reform.  We need to listen to what they have to say and focus on the removal of obstacles to good performance.  We need to think about what tools we can give them to make them better performers. Despite these criticisms, there is hope.  There is a push for a different type of government, focused on achieving results and holding employees accountable for the delivery of quality services to the American public.  

There is an approach to management that will work.  The President’s Management Agenda specifies the five ingredients of a high-performing agency. Performance budgeting means that government agencies need to justify their expenditures. Human capital planning requires assessment of the skills needed to get the job done, and anticipation of changes in agency human capital needs that will occur in the future. Competitive sourcing forces government managers to decide on making or buying goods and services. The PMA emerged when Bush asked what tools the line managers needed to be better performers.  He realized that the problems in government did not originate with the managers, but that they result from a system that is broken.  These five management reforms are not new.  They are consistent with statutes already in place.  What Bush has done is push for integration and accountability to drive good government performance.  He is advocating that we not treat these five areas as silos because all five goals are inter-related and tied together by performance.

Performance management in government is important, but that doesn’t mean that it’s easy.  Over the past 25 to 30 years, several different approaches to management reform have been launched, without realizing much success.  One of the comments often repeated is how hard it is to effect change in government.  What makes government managers and their staff members so resistant to change?  Inevitably, you will encounter resistance when taking the material contained in this presentation back to your agencies.  But why? 

Currently, the mentality among government managers is to not be too  concerned about performance.  How many of you have heard the term “year-end money?”  This quote is a good depiction of the difference between the public and private sectors.  From August 1st to September 30th, government managers are on a mission to spend everything left in the budget to safeguard their budget allocation for the next year.  The evaluation of program performance does not enter into this process.  No one is worrying about results.

Peter Drucker’s quote sets a different tone, and supplies the rationale for performance measurement.  It challenges government managers to see things in a different light.  “What gets measured, gets done.”  When accountability is introduced into the system, people respond with better performance.  And, measuring just a few things motivates employees to work toward good performance specifically in the areas of focus. Performance measurement systems promote consideration of performance in the budgeting process and facilitate a transition from measuring what gets spent to what gets accomplished.  Performance measurement is a movement toward a goal-oriented government. 

Some government agencies have been trying to do this for years, but have failed because they’ve been doing it wrong.  Performance measures need to supply the kind of information that enables government managers to make good decisions that effect change.  The following list of pitfalls sheds light on where performance measurement systems used in the past have gone wrong: 1) Most government agencies are not hurting for performance measures.  In fact, most who have implemented them have way too many.  Systems with hundreds and even thousands of measures overburden managers without providing helpful insight. 2) All too often, government managers fail to focus on the desired end result, but concern themselves just with measuring what can be achieved in the near-term.  Outputs, workloads, and activities serve as their performance measures.  It is important to move beyond simple activity and workload measures and be strategic about end outcomes and define measures that demonstrate real value to constituents. 3) Thirdly, they often focus on what is easily quantifiable, which doesn’t always lend itself to the most informative set of measures. 4) And finally, government managers often resist focusing on end outcomes and even intermediate outcomes because these measures make it hard to control or demonstrate progress on an annual basis.

In addition to these pitfalls, let me add an additional caution.  The process that we are talking about here is not just about defining measures.  The measures are merely the tools to apply to reach your goals.  It is essential that you survey your agency’s systems for planning, evaluation, management, budgeting, and oversight.  All of these play an important role in developing and implementing a system for performance measurement.

We have established the importance of what Peter Drucker was talking about and established that “what gets measured, gets done.”  Now where do we go from here?  What should guide our selection of performance measures?  How can this system prove most useful to program managers and their staff members?

For the answers to these questions, I turn your attention to the logic model approach – the heart of this presentation.  In order to promote good performance in government, we need to start with a logic model that links the activities of various departments and bureaus to well-defined strategic end outcomes consistent with agency mission and mandate.  The logic model is a proven approach for tackling the problems in government that we have discussed.

First, what is a logic model?  And why is it helpful for motivating and measuring government performance? The logic model is “a logical chain that links events and provides a blueprint for mission achievement.” The logic model is a graphic representation illustrating the rationale behind each program.  It’s not a 20-page glossy strategic plan, but rather a visual depiction of the causal relationships among today’s activities, future end outcomes, and the strategies in between. It is goal-oriented, containing the goals and performance measures for each phase of a program’s life cycle. And, it facilitates the planning, management, and evaluation functions of an agency to provide the big picture for agency operations.   

This diagram illustrates how the Government Performance Logic Model ties everything together and delineates the role of each program in the achievement of agency goals.  As the top arrows indicate, the first step is to determine the desired end outcomes and to define what success in those areas would look like.  Working backward from these end outcomes and their accompanying measures, you determine strategies for their achievement and then define the workload activities and necessary inputs of human and financial resources.  Once the outcomes, outputs, and inputs for the agency’s mission-oriented programs have been specified, the role that each of the function-oriented programs play in the achievement of agency goals needs to be determined. The five-year strategic plan contains outcomes and strategies.  Government agencies craft annual performance plans based on their budgets and annual operations plans.  The human capital plan ensures the employment of a workforce capable of carrying out the activities specified in the logic model.  Competitive sourcing and contracting deal with the decision to make or buy, and the e-government plan indicates the ways in which IT can assist in the accomplishment of agency goals.  Performance budgeting addresses how to pay for all of these organizational activities and outputs, and improved financial management ensures cost-effectiveness.  SES performance plans hold managers accountable for the specific goals they are charged with achieving.  And, the accountability and performance report ties everything together.

You may be interested in learning that the logic model traces its origins to the field of behavioral psychology where it provided the methodology for effecting change in human behavior.  This strategy was adopted by the private sector as a means to sell their products by appealing to the wants and desires of customers, in addition to members of the academic community involved in program evaluation. Non-profit organizations and public sector agencies embarking on the development of systems for performance measurement have begun adopting the logic model.  Currently, this methodology is being refined and implemented by The Performance Institute to promote good performance in public sector agencies.

You may ask why the logic model approach works when applied to government agencies.  It works because this methodology was designed by government managers for use by government managers. 

First of all, the logic model approach is outcome-oriented, which means that you work backward from your end goals.  The starting point is consideration of agency mandate and examination of the services that your agency is  charged with delivering to the American public.  Once you have set the desired end goals, you need to consider the changes necessary to achieve them, and then determine the best ways to measure these strategies and goals.  Let your managers figure out the activities and outputs that contribute to the accomplishment of these outcomes. It is also policy-relevant, which means that the logic model assists you in conveying information to stakeholders and policymakers.  It is macro measurement for macro management.  Performance measures enable you to gather the information you need in order to convey to Congress the implications of the statutes that they have mandated.  This communication with policymakers requires that you measure what is strategically important.

Then there’s the benefit of program integration.  Performance measures aid in the correction of the silo effect, whereby program managers work independently of one another.   Common measures of performance tie together different organizations at the federal, state, and local levels. There is also the benefit of systems integration, whereby performance measures ensure that the various departments of the agency work together for better management.  For instance, the HR, IT, budgeting, planning, and evaluation departments of an agency coordinate their operations. Finally, performance measures introduce accountability.  Once performance measures for overall agency functioning are developed, the next step is linking them to individual employee plans so that everyone is cognizant of the role that they play in ensuring the agency’s good performance.  

This slide contains vocabulary that is quite familiar to all of us as a result of GPRA – input, output, and outcome.  An input is the “amount of resources devoted to a program activity.”  With regard to a job training program, the inputs are the financial and human resources – 1 million dollars and 50 FTE employees. Your outputs are quantitative and qualitative measures of activities – your numbers of program participants and training courses offered. And finally the outcomes – the “assessment of the results of a program activity compared to its intended purpose.”  This program is intended to train people to land jobs.  The percentages tell us the pool of people who secure jobs as a result of the training.  Performance measurement can be done better than this.  By starting with the inputs and outputs, we have locked ourselves into preordained strategies and end outcomes.  But, what if this training program in its current form isn’t really the way to go?  By starting with it, we have closed the door to other potentially  more effective strategies.  The method depicted in the diagram above is too concentrated on three separate buckets – inputs, outputs, and outcomes.

Let’s continue with the example of the job training program, but focus on working backward from the end outcomes.  We need to ask ourselves, “What is the desired end outcome of a job training program?  Why do these programs exist?”  We need to get to the heart of the tangible results we are charged with delivering to the American people. I would posit that an important end goal of these programs is the achievement of self-sufficiency by program participants by earning a living wage. This is a different end goal from the previous slide where the goal was program participants landing jobs.  On the previous slide, we didn’t concentrate on the importance of participants earning a living wage.  The focus on this outcome limited our options in terms of activities and outputs.

The following list of questions should guide your specification of end outcomes.  Toss these out when you are in the role of facilitator at the stakeholder sessions to get the dialogue going. The next question, “How will you know you have been so successful that you can shut your program down?” is among the most intriguing on this list.  Allocate a good amount of time for stakeholder session participants to ponder this one.  Ask them to envision that point in time where the social problem that their agency is charged with addressing has been eradicated.  What would that look like? Next, what 2-3 pieces of evidence would you present to the grand jury (i.e. Congress) to prove your program’s worth? 

And, here are the questions to consider during the stage in the process where we specify intermediate outcomes or strategies: What must change in the status quo to create the conditions necessary for goal attainment?  Who are the targets of change and what must they do?  These deal with the necessary attitudinal and behavioral changes. What are the current barriers to the fulfillment of these end goals?  What causes these outcomes not to be achieved? And the magic wand concept:  What do you wish for this agency and the population it serves?  Intermediate outcomes are synonymous with strategies. Let me explore the definition of strategy in more detail since it is so important to what we are talking about.

Now let’s move to a focus on the specification of outputs and activities, which function as the means to achieve the intermediate and end outcomes.  We need to assign activities for the program managers to manage.  An example of an activity for the managers of a job training program would be to survey employers to determine the skills they look for in employees.  This activity would lead to the development of a curriculum delivering the skills and competencies sought by employers.  The output would be welfare recipients trained in the necessary skills to secure employment.  An example of an accompanying measure would be the percentage of program participants acquiring skills applicable to the jobs to which they have applied.

Toss out the following questions to generate dialogue among meeting participants during the development of program outputs to correspond to each intermediate outcome. At this point in the process, don’t concern yourself with administrative items.  Administration feeds all logic models. 

The most straightforward piece of the logic model is the inputs – the financial and human resources needed to deliver agency programs and services.  For the last step, we need to evaluate the amount of money and FTE employees needed to maintain the programs that we have developed and to fulfill the end goals we specified at the beginning of the process.

Here is an example of a performance logic model specifically for welfare-to-work.  Mind you, this is not a complete depiction of all of the inputs, outputs, and outcomes involved, but rather provides an example of how it works when applied to a specific program. Note the strategies or intermediate outcomes for achieving the end goals.  You see the attitudinal, behavioral, and conditional modifications necessary for reductions in total welfare benefits paid and the unemployment rate and an increase in the number of self-sufficient individuals. To the left are the necessary inputs and outputs – training courses, program participants, child care and transportation options, and the necessary human and financial resources.

Here you see a blank template for the construction of a logic model for your own agency’s programs.  Just as we did in the one for the job training program, the place to start is to the right at the end outcomes, working backward past the intermediate outcomes, outputs, and inputs.  The beauty of this model is in its depiction of the entire process, start to finish.  It enables you as a facilitator to clue your stakeholders in to the whole picture.  This template also enables you to note the comments of stakeholders as you go along, letting them see the pieces of the puzzle come together. Ideally, in working through the logic model for your agency, you want to bring together in the same room as many perspectives of the program as possible.  Invite program managers, outside stakeholders, policymakers, and so on.  Typically, meetings with program managers will occur separately from those with external stakeholders. The task of completing this logic model worksheet will take successive passes, and require refinement at each stage.

I want to go through the ground rules for this process.  As a facilitator, you should review these with the participants in your stakeholder sessions.  And, just as I did earlier, you should lay the groundwork for this process by conveying a sense of urgency and the need for this. Now the ground rules.  The first two listed here are extremely important since these rules imply the cultural shift that you need to effect.  In the beginning of the process, you should not feel constrained by things that you can “control,” but rather focus on what you can “influence.”  This should be a kind of brainstorming session.  Get everything down on paper before eliminating things and refining your list of end goals.   Other rules to convey to participants in the stakeholder sessions are the importance of measuring the real picture, without getting hung up on how politically sensitive the task is.  Also, do not feel constrained by the budget in the first go-around.  The list of outcomes, strategies, and activities that participants generate can be refined further down the line.  Finally, don’t start with activities or outputs, but instead start with end outcomes and work backwards.

This template aids in the development of the rationale or mission statement for each program and opens up the dialogue among participants at the stakeholder session.  Completing this form is a useful precursor to the construction of a logic model.  Referring back to the job training program discussed earlier in this presentation, let’s flesh out this template.  The mission of the Arlington County Job Training Program – is to provide practical education and training – to increase the competencies of welfare recipients – so that they can land well-paying jobs – resulting ultimately in them earning a self-sufficient wage so that they don’t depend on government benefits.

Here’s another helpful template.  Sometimes people define their activities as outputs.  The best way to deal with activities is to lump them into the specific output they serve.  This Activity Definition Template aids in clarification of the distinction between activities and outputs.  We start with work activities and cascade down toward intermediate outcomes.  Again, I will use the job training program example.  The purpose of case managers surveying employers and developing assessment tools and educational programs – is so that they can produce a training program – to build the skills and competencies of welfare recipients – so that program participants can obtain well-paying jobs.  

Now I want to apply the logic model specifically to the implementation of the President’s Management Agenda to show where this methodology really adds value to your agency’s focus on good performance.

But, before approaching the management reforms specified in the PMA, you have to examine program performance.  This template illustrates the functional areas of all government agencies.  It facilitates the process of determining how each of these functions – IT, human capital, financial management, competitive sourcing, and performance budgeting – support the achievement of an agency’s end goals.  Before addressing the PMA, government managers should develop logic models for each of their agency’s programs.     

In crafting your agency’s logic models, it is important to be deliberate about the selection of measures.  There are few perfect measures, and dealing with uncertainty in the area of performance measurement is inevitable.  In many cases, performance measures cannot tell the whole story, but they can provide red flags to indicate whether or not agency initiatives are successful.  Your focus should be on quality, rather than quantity.  Develop a few good measures, so that you don’t create a data burden for your program managers.  With the choice between 6 or 2 measures for a strategy, go with 2!  Put pen to paper to create a comprehensive list of indicators, and then refine the list from there by reducing it to those that are most useful.

Although there are few perfect measures, there are still several different types at your disposal.  This slide depicts a variety of measures. The baseline approach is very straightforward.  Today’s performance provides your baseline.  If your agency is effectively serving 800 clients on a monthly basis, your monthly caseload should only go up from there. Then there are leading, lagging, and current measures.  In the economic realm, you have leading indicators like the interest rate.  Lagging indicators, which by definition lag behind market action, include the average duration of unemployment.  And a current indicator is the GDP.  Benchmarking entails finding comparable programs against which you can measure the performance of your agency’s programs.  The results produced by a comparable welfare-to-work program shed light on the successes or failures of your own welfare-to-work program.

The Performance Institute has developed a set of criteria for the development of useful performance measures known as “SMART.”  The five words captured by this acronym should describe your system of performance measurement. Measurable – you have to actually be able to measure them.  Measures should be either numerically quantifiable or qualitatively definitive.  For example, a numerically quantifiable measure is reduction of the incidence of polio by 2%.  The eradication of polio is a qualitatively definitive measure. Accountability – You have to specify the point person or persons who are accountable for the measures. And finally, time-bound – you have to specify a time frame for the realization of your goals. I want to make these criteria more clear to you with two examples.  First, let me set the stage.  Suppose you leave this conference today and return home to your spouse, who is less than pleased and expresses her needs to you.  She really wants this marriage to work, but needs more attention from you.  The specific request is two hours of quality time without the kids per week.

This example embodies the “SMART” criteria I just presented to you.  She has laid out for you the intermediate outcome or strategy – two hours of quality time without the kids per week.  The end outcome that she wants is a happy marriage and no movement toward divorce.  Her demand is results-oriented, specific, accountable, measurable, and time-bound.  She has been specific – she needs quality time without the kids.  You will probably have to discuss further what constitutes quality in her eyes. Okay, how about one more example to make this crystal clear?  I want to focus on a reduction in smoking-related illnesses and death.  In order to facilitate the most practical, cost-effective use of resources, I want to select young people as my target population.  I want to intervene before they start or at least soon after they start. 

As I mentioned, the desired outcome is to reduce the incidence of smoking-related death and illness – and reduce costs.  My target population is young people, ages 10 to 18.  My measure for assessing success in this realm is to reduce the number of new youth smokers by 2% each year. This is results-oriented.  By targeting young people, hopefully before they start or even immediately after they have started, I have the best chance of making an impact.  The habit has not yet completely taken hold with this population. I have been specific.  I am focusing on the number of new young people smoking. Now you have seen two dramatically different applications of the “SMART” criteria as it pertains to performance measurement.

At The Performance Institute, we believe in the “learning by doing” approach to training.  Since this is the end of the presentation and you have learned all about the logic model, I encourage you to complete the program template in front of you for your own agency’s programs.  After completing this exercise, the methodology will become more concrete.  On another note, if you like what you have heard today, please feel free to contact us at the Institute so that we can come into your agency and assist you in the development of agency-specific logic models. I thank you for your attention today, and would be more than happy to answer any questions that you might have.