Module 13: Logic Models

A logic model is a visual representation of a program, showing how program resources are expected to lead to the desired health outcome or behavior change.  It itemizes all inputs, activities or processes, outputs, outcomes (short and long term) and impacts of the program, typically within a one page diagram.(1) The image below is a great resource for determining what programmatic information should be placed in each column. Not all logic models are set up in this format but most share the same categories of information; the flow and format depends upon the depiction that is most useful for the specific needs of the program implementers.
Description: C:\Users\ejenkins\Desktop\Reading-a-Logic-Model.png
Source: W.K. Kellogg Foundation, Logic Model Development Guide, p. 3.


These are the various people, physical items, and funding required to implement the program (i.e. human resources and staff, office space and supplies, or training materials).(2) For example, a program aiming to reduce diarrhea-related under-5 mortality rates may require 100,000 oral rehydration sachets (ORS), 10 community health workers (CHWs), and a local program manager to oversee the activities.


Activities listed in the logic model are the main substance of the health program.(3) For example, if the program is aiming to reduce diarrhea-related under-5 mortality in Ghana, one of the activities or processes may be "distribute ORS to village clinics" or "conduct educational sessions with mothers about how to properly administered ORS to children". 


These are indicators that should stem directly from the inputs and activities. In terms of timing, these indicators should be met by the time the program has ended.(4) Continuing with the under-5 mortality example, an appropriate output would be 100,000 ORS distributed to clinics in Village X in Ghana. This is accomplished by having the ORS available as a resource in the appropriate amount (100,000) and having one of the program activities to distribute the ORS; the result of this input plus activity is the output, or 100,000 ORS distributed. 


These are changes or improvements that can be measured among the target population or existing health system. These effects should be seen almost immediately at the end of a project implementation period for short term outcomes and, depending on the timeline and specific program characteristics, up to one to two years after for long term outcomes.(5) With the ORS example, an expected outcome would be a reduction in the incidence of acute diarrhea among the target population of children under 5. It is expected that, for one project, that there will be a number of outcomes but fewer than the number of outputs. For further distinction between outputs and outcomes, please refer to the Global Health Metrics module [] and the Outcomes in Global Health module [].


Impact is what the program implementers expect to happen, as a result of the program but after the program timeline has ended. Some organizations may refer to impacts as "long-term outcomes" but both essentially summarize intended health effects among  the target population or health system that occur several years after a program has ended and as program outcomes are achieved.(6) As with the ORS example, the impact would be to reduce under-5 mortality rates in Ghana. Most programs only have one or two impacts, often relating to reducing mortality rates or improving a health outcome, and they should link directly back to the program outcomes (reducing incidence of acute diarrhea will lead to a reduced child mortality rate).

Contextual factors

Although not directly related or planned by program developers, in any environment, there exist a number of contextual factors that will inevitably have an effect on the outcomes of the program. These may include political, social, cultural, or economic situations within the country or community that can create barriers, or sometimes promote the utilization of health care services or the adoption of new health behaviors.(7) The reason to include these relevant factors in the logic model is because they need to be measured to some degree to link outcomes to program activities and not external, uncontrolled factors.(8) For example, if there is a significant decrease in acute diarrhea, evaluators need to determine whether this was due to the effects of the program activities or whether income of the village was also increasing and therefore, allowing for income-related improved health outcomes. This analysis would involve collecting data on acute diarrhea incidence but also household or village income levels at various time points. 


Each of the components described should be recorded in bullet or list form in the logic model, not in narrative or paragraph form as the value of a logic model is in its simplicity. Typically, the logic model is accompanied in the report or grant proposal with a narrative explanation for any extra information that is of use but not vital to the logic model itself. The logic model is a useful tool for all stages of the program implementation and evaluation process and should be developed early on in these processes.(9) Like the first step, focusing the evaluation, developing a logic model or similar framework of the program pathways is important to include in an evaluation plan whether a process, outcome, or impact evaluation is being conducted.

Go To Module 14: Stakeholder Analysis >>


(1) W.K. Kellogg Foundation. (2004). Using logic models to bring together planning, evaluation, and action: Logic model development guide. Battle Creek, MI.

(2) McCawley, P.F. (n.d.). The logic model for planning and evaluation. University of Idaho.

(3) Ibid.

(4) Ibid.

(5) Issel, L.M. (2009). Health program planning and evaluation: A practical, systematic approach for community health (2nd ed.)  Sudbury, MA: Jones and Bartlett Publishers.

(6) Julian, D.A. (1997). The utilization of the logic model as a system level planning and evaluation device. Evaluation and Program Planning, 20(3):251-257.

(7) Bamberger, M., Rugh, J., and Mabry, L. (2006). Real world evaluation: Working under budget, time, data, and political constraints. Thousand Oaks, CA: Sage Publications, Inc.

(8) Victoria, C.G., Schellenberg, J.A., Huicho, L., Amaral, J., Arifeen, S.E., Pariyo, G., Manzi, F., Scherpbier, R.W., Bryce, J. and Habicht, J-P. (2006). Context matters: Interpreting impact findings in child survival evaluations. Health Policy Plan, 20(suppl 1): i18-i31.

(9) W.K. Kellogg Foundation. (2004). Using logic models to bring together planning, evaluation, and action: Logic model development guide. Battle Creek, MI.