Review Matrix

Worked example · Sources checked September 8, 2026

Literature Review Matrix Example

Review question: How does active learning affect learning outcomes and equity in undergraduate STEM, and how does students’ perceived learning complicate the picture?

These three real papers illustrate how to compare evidence. This is a small, deliberately selected teaching example, not an exhaustive search or a systematic review. The notes below paraphrase the linked abstracts and figure captions. Contribution labels, strength ratings, and scope cautions are editorial appraisals, not ratings assigned by the papers’ authors.

Download completed matrix (CSV)Download example backup (JSON)

To explore it in the tool, download the JSON file and choose “Import backup” in the generator. Back up your current draft first: importing replaces it. Then open the outline builder. For your own project, start with a blank template.

Source 1 · Freeman et al. (2014)

Active learning increases student performance in science, engineering, and mathematics

Method
Meta-analysis of active learning versus traditional lecturing
Sample
225 undergraduate STEM studies; 158 assessed test performance and 67 assessed failure
Key finding
Active-learning groups averaged 0.47 standard deviations higher on examinations and concept inventories.
Evidence location
Abstract; Figure 2 (discipline) and Figure 3 (assessment type and class size)
Theme
Learning outcomes
Contribution & appraisal
supports; strong for this example’s question
Limitation / scope note
Scope caution: a pooled effect across different STEM courses and assessments does not predict the gain in one course.

Source 2 · Deslauriers et al. (2019)

Measuring actual learning versus feeling of learning in response to being actively engaged in the classroom

Method
Random assignment to active or passive instruction with identical content and handouts
Sample
Students in large-enrollment introductory college physics courses
Key finding
Students learned more with active instruction but reported a lower feeling of learning than peers receiving passive instruction.
Evidence location
Abstract: controlled comparison of actual learning and self-reported learning
Theme
Learning outcomes
Contribution & appraisal
complicates; medium for this example’s question
Limitation / scope note
Scope caution: introductory physics and perceived learning are a narrower question than long-term retention or all STEM teaching.

Source 3 · Theobald et al. (2020)

Active learning narrows achievement gaps for underrepresented students in undergraduate science, technology, engineering, and math

Method
Meta-analysis with Bayesian regression of achievement gaps
Sample
15 examination-score studies (9,238 students); 26 failure-rate studies (44,606 students)
Key finding
Average examination-score gaps narrowed by 33% and passing-rate gaps by 45%; gap narrowing depended on the intensity of active learning.
Evidence location
Abstract; Figure 1 (average gaps) and Figure 4 (active-learning intensity)
Theme
Equity and implementation
Contribution & appraisal
supports; strong for this example’s question
Limitation / scope note
Scope caution: relative gap reductions are not percentage-point gains for individual students; implementation varied across courses.

Across the rows

From a matrix to a synthesis

The example distinguishes three outcomes: achievement, perceived learning, and achievement gaps. Keeping them separate prevents a review from treating satisfaction as evidence of learning or a reduction in a group gap as a guaranteed benefit for every student.

1. Learning outcomes and perception

Freeman et al. provide a broad comparison of test outcomes. Deslauriers et al. complicate evaluation: feeling that a lesson was effective and demonstrating learning need not align. These results address different measures and should not be described as a contradiction.

2. Equity depends on implementation

Theobald et al. extend the question from average performance to gaps between groups. Their intensity analysis makes implementation a useful theme for the review, rather than treating every activity as an equivalent intervention.

3. Bound the conclusion

These notes support a focused discussion of undergraduate STEM. A review about another population, online teaching, or long-term retention would need an additional search. Do not add the student counts across these different analyses as if they were one independent sample.

Why these appraisals? The two meta-analyses cover multiple studies and are marked strong for their stated outcomes. The physics comparison is marked medium for this broader review question because its setting is narrower. These are transparent reading judgments, not a validated risk-of-bias assessment.

Build your own matrixGet the blank template