Program theory or logic explains how and why a program is expected to work. It makes explicit the connection between the things we do in our programs and the outcomes we hope to achieve. Ideally, this logic should be evidence-based. That is, we should seek to consult theory and empirical research when developing a program theory.

The Council for the Advancement of Standards in Higher Education (CAS) recognizes the importance of program theory.(CAS)  recognizes the importance of program theory. According to their standards, programs should be 1) intentionally designed and 2) guided by theories and knowledge of learning and development.

A theory-based program is one that is intentionally designed to address a certain problem or achieve a certain goal. Theory-based programs can be contrasted with less intentional programs that are often conceived with little thought as to 1) what changes are desired in students' knowledge, attitudes, skills, and behaviors, and 2)  how such changes can be reasonably facilitated (i.e., program theory).

Developing a Theory-Based Program in Five Steps

Program Theory in Ac

Step 1: Articulate the Problem

Building a theory-based program begins by clearly articulating the problem you are trying to address or the goal you are trying to achieve. These problems/goals should be meaningful to stakeholders.  

Step 2: Identify Plausible Causes and Solutions

Once you have a clear problem or goal in mind, the next step is to consult relevant theory and empirical research to learn more about the underlying causes of the problem and theory-based solutions/strategies given those underlying causes.

Step 3: Select a Course of Action

There is likely more than one theory that has garnered empirical support, which is advantageous as it indicates more than one approach to designing a program to meet the intended SLO. Thus, the next step is to select a reasonable course of action given what you know about your specific institutional context (e.g., the unique needs of your students, resource constraints, institutional priorities, etc.).

Step 4: Develop an Intentional Program

Once you have decided on a course of action (i.e., what you're going to do), the next step is to figure out  how to do it. In other words, while theory may provide an outline for your program, it is up to you to fill in the details. A good starting point is to look for  intervention studies, or empirical research articles where practitioners describe a particular theory-based intervention and evaluate its effectiveness. It may also be helpful to consult the literature on how students learn (learning theory) and/or how attitudes and behaviors are changed (persuasion theory). Additionally, there is extensive literature on evidence-based pedagogical techniques that can be particularly useful for educational or co-curricular programs.

For additional guidance on finding evidence-informed programming, please visit Evidence-Informed Programming.

Step 5: Evaluate Program Effectiveness

Once you've built and implemented a theory-based program, the next step is to evaluate its effectiveness. If you find that your program is ineffective, it may be an indication that 1) the theory/research you pulled from in Step 2 is wrong or does not generalize to your population of students, 2) the course of action you selected in Step 3 is inappropriate given your specific context, or 3) the particular methods you employed in Step 4 were ineffective or insufficient. Once you've identified where the breakdown occurred, the final step is to make intentional, theory-based changes to the program based on this information.

Click here to see each of these steps applied to a hypothetical program targeting binge drinking behavior.

¹American College Personnel Association & National Association of Student Personnel Administrators (2015).  ACPA/NASPA professional competency areas for student affairs educators.Washington, DC: Authors.

A logic model is a depiction of the relationships between program components and outcomes. In other words, a logic model shows which program components will lead to which outcomes, and most importantly, how (i.e., through what causal mechanisms) this change will occur. In this way, logic models can be conceived of as illustrations of the program theory underlying a program. As such, the arrows in these models should be evidence-based.

Parts of a Logic Model

Logic models often contain some (or all) of the following components:

Click image to enlarge

A Realistic Example

The logic model below was created by Jeremy Penn in the Office of Teaching and Learning at North Dakota State University.

Click image to enlarge

The primary purpose of a program-to-outcome map is to ensure that each SLO is sufficiently addressed by your programming. In other words, does your program provide students with an opportunity to achieve all of the stated outcomes?

Additionally, program-to-outcome maps help with use of assessment results. If you find, for example, that students did not achieve a certain SLO, you can use a program-to-outcome map to figure out what components of the program need to be reevaluated.

Example 1: Simple Coverage Map

In this map, an "X" indicates which program components contribute to the achievement of which outcomes.

Check For Unnecessary Program Components

• Note that the first program component (video and small group discussion) does not map to any of the program's outcomes. This should prompt facilitators to reevaluate this activity. Is it necessary? Does it detract from other more important activities? Although there are some instances where it may be appropriate to include an activity that does not map to a specific outcome (e.g., some introduction or "ice-breaker" activities), there should be a compelling reason to do so.

Example 2: Degree of Coverage Map

This map is more detailed; it not only specifies which program components map to which outcomes, but the degree of coverage that each program component provides.

Check For Unaddressed Outcomes

• Note that Outcome 4 has minimal coverage. Only one component (i.e., the "Risk Reduction Plan" activity) is mapped to it, and the coverage of this activity is slight. This should prompt facilitators to question whether the program adequately equips students to achieve Outcome 4. Is additional programming necessary? Or should existing programming be strengthened?

Program Theory

As you refine your program theory, AI tools can help with brainstorming, clarifying connections, and organizing ideas; but remember that expert judgment is essential.

What do relevant theories say about how to achieve your SLOS?

For new programs:

Copilot Chat and ChatGPT can assist in:

• Refining preliminary descriptions of your program theory based on your SLOs and supporting theories.

!!! Note: Copilot Chat and ChatGPT should not be used to search for existing theories, they can provide false citations or incorrect information. They may be used to assist your work but should not replace expert judgement. !!!

Scopus AI can:

• Create a summary of theories that tie your programming to your outcomes with real citations and an associated reference list.
• Curate a narrow list of empirical articles that are relevant to your programming and outcomes.
• Build a concept map that ties concepts together in your field of study.

!!! Note: Avoid relying on the summary from Scopus AI as the sole text for your program theory. This summary should be used as a starting point to help you understand some of the different theories that support your programming and outcomes. You should conduct your own literature review to ensure all relevant literature has been consulted. !!!

For inherited programs:

If you’re working with an existing program theory, you can…

• Upload the theory to Copilot Chat or ChatGPT to check for clarity and logical flow.
• Ask AI tools to help identify missing links or unclear connections between components and outcomes.

Program Mapping

What programming would you implement to foster desired learning and development?

For new programs Copilot Chat and ChatGPT can assist in:

• Suggesting activities that align with your desired learning and development outcomes.
• Helping structure and build out your program components.

Why should this programming impact the intended outcomes?

For new programs:

AI can provide basic rationale for how your proposed activities align with your outcomes. Copilot Chat and ChatGPT can assist in…

• Clarifying why specific activities contribute to student learning.
• Ensuring your programming has a clear purpose tied to your SLOs.

!!! Note: AI will find a connection even if the connection is not strong or relevant for your specific programming. It may be used to assist your work but should not replace expert judgement. !!!

For inherited programs:

You may be given a program that already has program theory, components, and outcomes drafted. Tools like Copilot Chat and ChatGPT can help identify connections between program components and outcomes when you…

• Upload the relevant program theory, components, and outcomes into Copilot.
• Ask for the connections between the components and outcomes based on the uploaded program theory.

!!! Note: AI may produce hallucinations (evidence that is not real). It may be used to assist your work but should not replace expert judgement. !!!

What evidence supports the effectiveness of this programming?

For whom is this programming effective? Should it be equally effective for all students?

For new programs:

While Scopus AI can help you locate research supporting the effectiveness of your programming, generative AI tools like Copilot Chat and ChatGPT are not suitable for compiling evidence. Though collection of evidence for your programming should be collected prior to this step of the assessment cycle, you may use Scopus AI for…

• Finding relevant studies that show how similar programs have worked.
• Collecting evidence on what makes certain activities effective for your targeted outcomes and student population.

!!! Note: Avoid using generative AI for justifying effectiveness across diverse populations, as it can’t fully address context-specific equity and limitations. !!!

BONUS TOOL

One of the most practical uses of AI in this step is creating program maps. These maps intend to connect your program’s distal outcomes, intermediate outcomes, and program components to make an easy-to-follow visualization of your programming.

A tool like Eraser’s AI Flowchart Generator lets you:

• Input your various outcomes, components, and evidence-based connections in simple conversational text
• Automatically generate a visual program map
• Make easy edits as needed

Tip – This tool is most effective when you’ve already outlines your outcomes, theories, and components. And you can sign in with Microsoft (admin approval required) or Google to save and edit your maps.

DO CONSULT RELEVANT THEORY

Reading the research in relevant domains can help you to create SLOs that are not only specific, measureable, and reasonable, but also evidence-based. Imagine you want to develop a program to increase students' civic engagement. You might specify that upon completion of your program, students will be able to list three elected state officials. But does knowledge of state officials actually lead to increased civic engagement? Without an understanding of the research on civic engagement, you are likely to write misguided SLOs that have little hope of truly bringing about the impact you desire. Thus, theory should be consulted before specifying SLOs.

DO USE INTERMEDIATE AND DISTAL OUTCOMES

Often the impacts we wish to have on students—the things we truly care about—are hard to define and difficult to reduce to measurable outcomes. Thus, specifying distal outcomes may prove a useful way to start the outcomes writing process. Distal outcomes capture our more general aspirations and guide the development of specific, intermediate outcomes. The intermediate outcomes are not meant to fully articulate the breadth of the broader distal outcome; there are often a multitude of specific intermediate learning outcomes that can be mapped to a distal outcome. The intermediate learning outcomes articulate how progress toward the distal outcome will be evidenced.

DO ARTICULATE OUTCOMES BEFORE CREATING PROGRAMMING

SLOs serve as the foundation for program development; it is nearly impossible to intentionally build programming to impact particular outcomes without stating those outcomes first. In short, outcomes are stated and then programming is built that should (according to theory) impact those outcomes. Programs built without explicitly-stated SLOs can be immensely creative and fun, but it is extremely difficult to assess the effectiveness of these programs without first figuring out what outcomes they may impact and why. As a result, students may not experience a sufficiently targeted intervention that increases learning or development.

DO INCORPORATE OTHERS INTO THE PROCESS

For SLOs to be meaningful, they should encompass the voices of as many stakeholders as possible—including those who will develop, implement, oversee, and experience the program. Moreover, you may have colleagues on campus who are working toward similar outcomes and who can provide insight into the specification of outcomes and the development of programming (there may even be partnership opportunities). When one person develops SLOs with little or no input from others, the result may be outcomes (and by extension, programs and assessment results) with limited value to others or redundancies across the division.

DON'T RUSH THE PROCESS

It is not uncommon for the process of developing meaningful, measurable SLOs to take weeks or months of dedicated effort. Remember, your SLOs are the foundation for everything else you do. Before writing them, take time to consult theory, incorporate stakeholders, and think critically about what students can reasonably accomplish in an allotted amount of time.

The ABCD method is a great tool to assist in writing clear SLOs. "ABCD" is an acronym that refers to four important components of any SLO: audience, behavior, condition, and degree.

An example of an SLO written using the ABCD method:
As a function of living on campus (Condition), the first-year residence hall students (Audience) will develop a greater sense of belonging (Behavior) as indicated by a 2-point increase on self-report instrument (Degree).