ABET Criterion 4 examples: what a closed loop actually looks like

What Criterion 4 asks for

Criterion 4 turns on two things the questionnaire keeps separate. Assessment is how you gather data on whether each student outcome is being attained. Evaluation is how you interpret that data to decide how well the outcome is met. Then, the part that carries the most weight, you show how those evaluations were used as input to actually change the program, and whether the change worked. In the questionnaire, section 4.A asks for the assessment and evaluation; section 4.B asks for the improvement actions and their results.

The anatomy of a strong Criterion 4 example

A complete continuous improvement example has all of these parts. Where one is missing is usually where a shortcoming gets written.

• The outcome. Which student outcome is being assessed.
• The measure. The specific instrument: a capstone rubric, an embedded exam question, a portfolio review, an internally developed exam.
• The frequency. How often the outcome is assessed.
• The target. The expected level of attainment, stated as a threshold before you look at results.
• The result. What the data actually showed, with a summary and analysis.
• The evaluation. Your interpretation: is the outcome being attained, and if not, why.
• The action. The change made to the program in response.
• The re-assessment. The follow-up showing whether the action moved the number. This is the step that closes the loop, and the one most often missing.

An illustrative example

Here is a simplified, generic loop. It is illustrative, not from any specific program. Outcome: the ability to apply engineering design within realistic constraints. Measure: a capstone design rubric scored by two faculty. Frequency: annually. Target: 75 percent of students at proficient or above on the requirements-analysis dimension. Result: 58 percent, below target. Evaluation: faculty found students were jumping to solutions without defining constraints, traced to thin coverage earlier in the curriculum. Action: added a requirements-analysis module to a junior-level design course. Re-assessment: the following cycle, the same rubric dimension rose to 76 percent. That final step is what makes it a closed loop rather than a report of a measurement. A reviewer can see the outcome was assessed, evaluated, acted on, and re-checked.

Where Criterion 4 examples fall apart

The most common failure is not a program that skips continuous improvement. It is a program that does the work but cannot show it on paper. The assessment happened in one faculty member's spreadsheet. The decision to change a course was made in a meeting whose minutes were never filed. The re-assessment exists but nobody connected it back to the original action. On paper the loop is open, even though in real life it closed. Criterion 4 asks you to make the loop legible across a six-year cycle, and that is a recordkeeping problem, not an improvement problem.

The modern way to keep Criterion 4 ready

This is where dal.ai fits. The continuous improvement module structures your work as the closed loop Criterion 4 expects: outcomes, measures, targets, results, actions, and re-assessment, kept together over time rather than reconstructed the year before a visit. When it is time to write the self-study, dal.ai drafts the Criterion 4 narrative from that record and cites the source behind every claim, and where a loop is still open it flags it so you can close it before a reviewer does. You review and approve. You are still the expert. It is your system of record.

Frequently asked questions