3.9 What is the difference between 4 Quadrant performance measures and logic model performance measures?

The Short Answer

1. The 4 Quadrant method and logic model methods can be seen as complementary, not contradictory, approaches.

2. The 4 Quadrant model goes directly to the identification of performance measures, without a lot of preliminaries:

  • What do we do? (quantity of effort = # clients served, # activities performed) (= logic model outputs).

  • How well do we do it? (quality of effort = % timely actions, % complete actions, unit cost, etc.)

  • Is anyone better off? (quantity and quality of effect = # and % of clients who show improvement in skills, knowledge, attitude, behavior or circumstance) (= logic model outcomes).

3. There are many different frameworks using the term “logic model.” Most logic models take time to describe program components in writing (sometimes in great detail) and derive client outcomes from the following sequence: 

  • Inputs (usually resources like money, staff, space etc.).

  • Processes (what activities the program performs) .

  • Outputs (what units of work the program produces) 
    (= quantity of effort – upper left quadrant). 

  • (Client or customer) Outcomes (what effect the program has on the lives of its clients)(= client results or outcomes – the lower quadrants). Sometimes this is a two step process which states client outcomes in plain language and then in terms of data statements. 

  • Goals (the social condition to which the program contributes)(= population results or outcomes).

4. The corresponding component of each is highlighted. Many logic model approaches do not address “quality of effort” measures and these must be added be separately.

5. The identification of performance measures, using the 4 Quadrant model, a logic model or any other model, is part of a larger process. It is the starting point for using performance measures to improve performance. 

See 3.14 What do we do with performance measures once we have them? How can we use performance measures to improve performance?>See also QA1.3  Logic Model � Results Decision making: Which process to use?

Full Answer

Click here for a Crosswalk from the Results-Based Accountability framework to the logic model framework

Logic Model and Theory of Change methods are related and sometimes lumped together. The following section will address them separately. Logic models are most often used for evaluation of programs. Theory of Change is a more broadly based approach that can be applied to evaluation of both programs and community-wide efforts to improve the well-being of a whole population. 

(1) What is a “logic model?”: Logic models provide a method of describing the way a program is intended to work. This is presented as a logical sequence 

  • from inputs (usually resources like money, staff, space etc.).

  • to processes (what activities the program performs) .

  • to outputs (what units of work the program produces) – this corresponds to quantity of effort – upper left quadrant). 

  • to outcomes (what effect the program has on the lives of its clients – this corresponds to client outcomes or results – the lower quadrants). 

  • to goals or community impacts (what contribution the program makes to larger population conditions – this corresponds to population results).

(2) There is a great value in the work being done under the rubric of “logic model,” or “theory of change.”  Logic models and the Results-Based Accountability model presented in this guide are all, of course, “logical.” And in fact both approaches are complementary. But there are important distinctions worth noting: 

(a) Logic model work is mostly about program performance and generally does not address whole population well-being. It is a performance accountability method or program evaluation method. 

(b) It works best for programs which provide a specific service or set of services. (It is often used to design program evaluations.) It is less useful for whole agencies or for parts of agencies that do not provide service (like administrative functions). 

(c) Logic model thinking processes, like the United Way’s approach,  work in the “opposite direction” of results-based decision making and budgeting. One starts with clients, the other ends with clients. 

Logic models start with program resources and then describe activities,  work product outputs and finally the ways in which clients are better off, i.e. client results.

Performance accountability skips directly to “who are our clients?,” then identifies client results, baselines, the story behind the baselines, partners, what works and an action plan.

These can be seen as complementary thinking processes. It would be possible to use a logic model to identify client results and then pick up with the performance accountability process from client results to action.

(d) Many program logic models do not address quality of effort (upper right quadrant) measures like staff ratio, timeliness, cultural competence etc. This must be added.

(e) Most logic models do not distinguish between the quantity (#) and quality (%) of client results. (“Our program helped get 15 people off of drugs.” This number is good if the total number served was 20; and bad if the total served was 1,000. The percentage or rate shows this difference.) 

(3) Logic models are particularly useful in designing programs, because they require you to think about the logic of how the program is supposed to work. 

(4) The logic model thinking process essentially answers the question “How is this program supposed to affect client results?” It is the kind of test people should use in the “what works” part of the results- based decision making framework. As such it is a tool that can help test  ideas about what works, and help make the best programmatic choices about how to turn the curve on client results.

(5) In some cases program staff have found it helpful to think through how a program is supposed to work (using some form of logic model) and then developing performance measures in the 4 Quadrant framework. This can sometimes help people who do not understand the concept of client outcomes and have trouble getting to the lower right quadrant measures of whether clients are better off.

See also 1.8 How do we fit together different approaches when there is more than one approach to Results-Based Accountability being used in my area?