Logic Models

A logic model identifies inputs, activities, outputs, and outcomes used to measure successful or achieved objectives; and serves as a blueprint for the implementation and evaluation of the program, strategy, service, or intervention studied.

To learn more about logic models visit pages 28-32 of the DOL Framework or view the WIOA Wednesday Webinar PDF Presentation and recording, “Evaluation Under WIOA:  Planning and Performing an Evaluation” from May 6, 2020, below. 

It is a good idea to begin the preliminary evaluation plan with a logic model. A detailed logic model may not be necessary, particularly if the focus of the study is narrow rather than comprehensive. Even if the study focuses on a program activity or strategy, a simple logic model delineates program inputs, outputs, activities, and outcomes.

Evaluation Under WIOA:  Planning and Performing an Evaluation- May 6, 2020

Presentation

This video includes a description of planning an evaluation and discusses the role logic models play in the process.

A logic model relies on a specific “theory of change.” Examining the logic underlying a program (i.e., program, system, strategy, service, activity, or intervention) clarifies the subject matter and context of the evaluation. For example, within a logic model, program implementation describes the inputs and outputs, and the program results are expressed as outcomes and impacts. The logic model description is a detailed account of the program’s content and organizational structure, size, flow, staff support, the amount of staff training required to implement it, and the services provided, or system change activities undertaken. It may also contain a clear depiction of the relationships between program elements and the intermediate- and long-term outcomes those elements are expected to affect. A well-designed logic model serves as a blueprint for the implementation and evaluation of the program it describes.

Maps program management operations—

identifies inputs, activities, outputs, and outcomes used to measure successful or achieved objectives; and

Serves as a blueprint for the implementation and evaluation of the program, strategy, service, or intervention studied.

Key components in logic models are inputs and activities used to operate a program. The logic model components are the ingredients in the program implemented to achieve the desired outcomes. The key components—essential activities and inputs—may include, for example, financial resources, professional development for trainers, curricular materials, or technology products. The inputs and activities describe the program in action and summarize the required operations to attain fidelity to the model. Regardless of the logic model is linear or nonlinear, at a minimum; it identifies the basic operational standards and structure in graphic form organized to demonstrate how the outputs lead to the program’s intermediate- and longer-term outcomes.

Outputs are products developed, deliverables completed, or milestones accomplished from the program activities and inputs. In other words, program outputs occur when the inputs and activities accomplish the intended objectives. Some examples of workforce program outputs include:

  • Participant services (e.g., skill assessments, occupational training) delivered;
  • Training/professional development activities or other supports for trainers (e.g., group training, on-site coaching, distance training, curriculum materials) completed;
  • Instruction or skill development (e.g., technology, formative assessment(s), use of instructional time, participant groupings) completed;
  • Participant, employer, and community engagement activities completed; or
  • Coaching, advising, or referrals developed and delivered.

Intermediate outcomes are the expected program service delivery milestones or goals achieved that can lead to long-term outcomes. A logic model includes all intermediate outcomes through which the program is expected to affect participant long-term outcomes. Note that outputs and intermediate outcomes of workforce programs are often the same. Some examples include:

  • Training program completions;
  • Participant credentials;
  • Participant job placements;
  • The number of overlapping services reduced; and
  • The number of complete records entered into a new management information system (MIS).

Long-term outcomes include the expected changes in behavior, attitudes, aptitude/skill, and knowledge for staff, participants, environments, or larger systems. Most importantly, workforce programs include changes in employment and earnings, employment retention, and the receipt of credentials as part of long-term outcomes. All outcome domains that the program is expected to affect should be included in the logic model.

  1. Does the model include critical inputs required for the implementation of the service activities? (e.g., accessible technology and other resources, program partner services, staffing, coaches, case managers, recruiting and training trainers, and partners with training services, credentialing, and work-based learning opportunities)
  2. Are there system-building activities that are part of the necessary foundation for the program? Are these system-building activities part of the logic model?
  3. Does the model include all current activities provided to participants? Is there an existing or expected sequence of participant activities that follows a logical path or pattern?
  4. Does the model include all “first-level outputs” of the program? (For example, measurable milestones that are necessary but not sufficient conditions for achieving outcomes, such as full participation, use of supportive services, and meetings with coach/advisor.)
  5. Does the model include all hypothesized immediate changes and/or outcomes expected for participants, across all relevant domains?
  6. Are these immediate changes and/or outcomes an assumed result of specific services?
  7. Does the underlying theory of the program design identify expected participant outcomes for particular services?
  8. Does the logic model suggest links between intermediate- and longer-term outcomes?
  9. Are the longer-term participant outcomes likely to be measurable in the life of the evaluation?
  10. As a complete visual or narrative text, does the logic model tell a clear and complete story about the unique program, service strategy, or intervention in the study?
  11. If the logic model assumes a theory of change, how does the hypothesis lead to moderate and long-term outcomes?
  12. If using a visual representation, does the supplementary narrative text provide a clear and complete story?
  13. Are there assumptions about external conditions or other external factors that could affect the successful implementation of the program?
  14. Are these identified external conditions or other external factors shown in the model?