Differentiated instruction using pre- and post-tests

Intent

Many equitable grading approaches de-emphasize or eliminate conventional ranking-oriented methods of assessment in favor of more individualized assessments that take into account students’ starting levels of preparation and growth during their time in a course. This play discusses an approach using pre- and post-tests to identify weak points in a students’ starting understanding of a topic, which can then serve as a guide to creating an individualized instruction plan.

Problem

Conventional assessment methods assume that students enter a course with roughly equal levels of preparation and then score and rank students against each other to reward the highest performers and remove the lowest performers.

Standard assessment methods may not be applicable if students enter a course with unequal levels of preparation. Further, a ranking-oriented approach with failure as the punishment raises student anxiety, which harms engagement and shifts students’ goals towards meeting grading expectations rather than learning. Some educators see traditional grading as reflecting a fundamentally capitalist orientation and would prefer to use an assessment approach that draws upon different values.

Solution

There are a variety of plays that may fall under the “ungrading” category. In this one, assessment is based on individual student improvement over the course, as measured by pre- and post-evaluations.

• At the beginning of the course or unit, administer a pre-test to establish each student’s current level of conceptual understanding
• Based on the results, provide differentiated instruction to individual students to improve their weak points as indentified by the pre-test
• At the conclusion, administer a post-test and evaluate each individual student’s growth relative to the baseline results

Applicability

The primary advantage of this approach is its move away from the “one size fits all” assumption of conventional grading. In theory, supplying differentiated instruction should help students learn more and faster, since they’re receiving focused engagement on weak points.

In practice, though, differentiated methods are hard to implement and scale. Instructors need to carefully consider how they’ll actually implement differentiation: unless the class is very small, one-on-one (one-to-few) tutoring is not viable. In most cases, this approach assumes some type of electronic resource that can be customized to each student’s weak points. This is discussed in more detail below. Instructors also need to select (or design) the pre- and post- assessments. Some model assessments exist in specific areas, like the Force Concept Inventory for physics, but there are relatively few such standardized assessments for computer science.

Unless an instructor choose to forgo conventional grades entirely, there’s still a need to map results of the pre- and post-assessments to actual scores or letters.

How to Implement

Here is a summary of recommendations included in the source:

• Select a concept inventory or other assessment that can be used as a pre- and post-assessment of students’ mastery of the content. This could be decided at the level of the entire course, a unit, or a single topic.

• Adminster the pre-assessment to every student.

• Analyze the scores to identify significant gaps and opportunities for improvement for each student.

• Develop a differentiated instruction plan with individualized recommendations for each student. The authors suggest that AI tools may be helpful in providing this differentiation, but don’t supply any specifics of how AI-based instruction could be implemented.

• Periodically, test students again on the assessment. Determine each students’ gain over the baseline level established at the beginning.

• Monitor and adjust instruction in response to each round of assessments.

The authors implemented this strategy in a physics course using the Force Concept Inventory as their assessment instrument. Students grades were determined by scaling the improvement scores to a fixed range, taking into account the distribution of improvement scores across the class. The grading system has to account for topping out: students who are already close to the top score on the assessment have less room to improve and are rewarded for maintaining a high score even if their improvement is small in absolute terms.

Source

Source: Crogman, H. T., Eshun, K. O., Jackson, M., TrebeauCrogman, M. A., Joseph, E., Warner, L. C., & Erenso, D. B. (2023). Ungrading: The Case for Abandoning Institutionalized Assessment Protocols and Improving Pedagogical Strategies. Education Sciences, 13(11), 1091.

Described by: Dan S. Myers, dmyers@rollins.edu

Community Discussion

Community members are free to comment on, ask questions about, share experiences, or otherwise contribute to knowledge about this play by posting comments below. See Chapter 33. Join Our Discussions for details.

• Insert a comment here.