The effect of color-coding on students' perception of learning in introductory mechanics

Brianna S. Dillon Thomas; Scott Carr; Siming Guo

REVIEW 3 major objections 5 minor 22 references

The effect of color-coding on students' perception of learning in introductory mechanics

T0 review · 3 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Color-coding is perceived as helpful in introductory mechanics slides, with a consistent minority of students noticing the schemes even when no attention is drawn to them.

desk verdict A careful descriptive survey of student perceptions of color-coding in mechanics slides; the abstract's '40% of students' overstates response-level data, but the core practical finding is likely robust. read the letter →

arxiv 2411.14605 v1 pith:I5RV3QTW submitted 2024-11-21 physics.ed-ph

Brianna S. Dillon Thomas , Scott Carr , Siming Guo This is my paper · ORCID

classification physics.ed-ph

keywords color-codingintroductorymechanicsstudentperceptionsmultiplerepresentationsslide-basedinstructionphysicseducationsurveyanalysis

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tests whether three color-coding schemes used in slide-based introductory mechanics instruction help students learn by helping them connect related representations and distinguish separate pieces of information. Using anonymous open-ended surveys across two semesters, the authors find that students' views are generally favorable: most responses identify something positive about color use, few criticize it, and a consistent minority of responses (about 16%) can describe a color-coding scheme as helpful even when it was never pointed out to them. About a quarter of responses found color in a physics context helpful without describing the color-coding itself, and on average 40% of responses said color helped them match or connect related information or separate and distinguish distinct information. The paper concludes that thoughtful color-coding is a worthwhile, low-risk instructional practice, especially when applied to mathematical variables paired with verbal or visual representations.

What carries the argument

The intervention consists of three color-coding schemes: (1) matching a word in a verbal definition with the corresponding variable in its mathematical definition; (2) matching a variable in an equation with the corresponding element in a diagram; and (3) using separate colors for x-components, y-components, and shared quantities in two-dimensional scenarios. The analysis rests on a two-part classification of open-ended survey responses: 'Type' categories capture how fully a response described a color-coding scheme (ranging from full color-coding recognition to generic color mention), and 'Reason' categories capture why color was said to help (matching/connecting, separating/distinguishing, tracking, comprehension, appearance). Classifying responses this way lets the authors separate students who consciously noticed the schemes from those who only noticed color, and to compare which schemes students found most helpful.

What would settle it

Administer the same survey items with linkable student identifiers and recompute the percentage of distinct students who say color helped match or separate information; if the on-average 40% figure drops below about a quarter, the paper's central proportion would not be supported.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that in a flipped, calculus-based introductory mechanics course, students perceive systematically colored slides as helpful. Across the five survey questions analyzed, 75.3% of responses identified something positive about the question-relevant use of color, 6.8% identified something unhelpful, and 2% were exclusively negative. Even without the schemes being named or pointed out, an average of about 16% of responses clearly described a relevant color-coding scheme as helpful, and a further 27% found color helpful in a physics context without describing the scheme. The reasons students gave aligned with the intervention's goals: on average 40% of responses said color helped them match or connect related information or separate and distinguish distinct information, which is comparable to the share citing generic comprehension or aesthetic benefits. Students most often favored color used on mathematical variables and in the equations-and-definitions (Scheme 1) and equations-and-diagrams (Scheme 2) contexts, and the few negative comments pointed to fixable issues such as too many colors or similar hues.

Load-bearing premise

The survey responses are anonymous and pooled across questions, so the reported percentages treat each response as an independent voice even though the same student could have answered up to nine questions; the paper itself notes that it cannot know whether the same students form the 'Color-coding' response group each time.

Editorial extensions

If this is right

Instructors using slide-based physics instruction can adopt color-coding at little cost, since very few students object and most criticisms are addressed by using fewer, more distinct colors.
Color-coding appears to help most when applied to mathematical variables and equations, not just diagrams; this suggests instructors should color mathematical expressions when linking them to definitions or diagrams.
Because favorable responses increased when students were prompted to think about color, explicitly drawing attention to a consistent color-coding scheme may increase its perceived benefit.
The paper motivates a follow-up test of color-coding in mathematics-only contexts, since many students found color on equations helpful even without a link to verbal or visual representations.
The consistent minority who recognized the schemes without prompting suggests some students spontaneously use color cues to organize information, while others need the cues pointed out.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the anonymous responses were linked to individuals, the 40% figure might shrink or shift, because the same student could have answered several of the nine surveys; the paper itself notes this unknown.
Perceived helpfulness is not the same as measured learning; a natural extension would compare exam performance in sections that differ only in color-coding.
The authors' reading of prior work suggests color-coding may help students with weaker visual-organization skills most, but this study does not test that subgroup claim directly.
The three schemes are cheap to implement in other STEM courses that rely on multiple representations, so the perception results may transfer beyond physics if similar surveys are run.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: a careful, honest survey study with modest but real practical value. It does not measure learning, and the title's 'effect' is not supported by the design. But the finding that a consistent minority of students spontaneously notice and value color-coding, and few object, is solid.

What's new: no prior PER study tests color-coding in intro mechanics. The authors designed three concrete schemes and collected open-ended responses across two semesters. Coding is detailed and consensus-based; they report inter-rater agreement and handle ambiguous cases transparently. That is genuine work.

Soft spots: the numbers. The abstract says 'on average 40% of students found color helpful in matching or separating,' but the denominator is responses, not students. Students answered up to nine questions, anonymously, so the same student can appear many times. The paper itself admits this in Section V.B, yet the abstract and conclusion slip into 'students' language. The 40% may also combine two overlapping categories rather than the union of responses endorsing either. Fix the wording to '40% of responses' or report a per-student bound.

The title overstates. 'Effect' implies a causal comparison; there is no control condition. They measured self-reported perception, which is fine, but the causal word should go.

The 'Reason' categories do overlap with the authors' stated motivations, but they emerged from the data and the appendix shows careful application. I see this as a minor circularity, not a fatal one.

Who benefits: instructors looking for practical slide-design guidance, and PER researchers working on representation and visual attention. Not a rigorous causal study; don't cite it as one.

My recommendation: send it to review. The data and coding are solid enough to publish after a careful revision of the unit-of-analysis language and a less causal title.

Referee Report

3 major / 5 minor

Summary. This paper reports a descriptive study of students' perceptions of color-coding in slide-based introductory mechanics instruction. The authors implemented three color-coding schemes (equations-definitions, equations-diagrams, and 2D-component separation) without explicitly pointing them out to students, and administered open-ended surveys at multiple points across two semesters. Responses were coded by three raters into 'Type' categories (degree to which a color-coding scheme was described and whether the opinion was helpful/indifferent/unhelpful) and 'Reason' categories (specific reasons such as matching/connecting information or separating/distinguishing information, plus generic reasons). The headline findings are that most responses were favorable, about 16% of responses described a relevant color-coding scheme as helpful, about 27% described color in a physics context as helpful without describing color-coding, and 'on average 40% of students' found color helpful for matching/connecting or separating/distinguishing information. The paper concludes that thoughtful color-coding is a worthwhile instructional practice and offers implementation suggestions.

Significance. If the quantitative claims are interpreted at the response level, this study provides useful descriptive evidence about how introductory mechanics students perceive systematic color use in instructional slides: a consistent minority spontaneously notices and values color-coding, negative reactions are rare, and specific implementation contexts (equations and diagrams) are especially favored. The paper's strengths include the use of open-ended questions rather than leading rating scales, a detailed and transparent coding scheme with reported inter-rater agreement, and the authors' care in distinguishing between students who recognized color-coding and those who only described generic color use. The study is however exploratory and small, and the headline '40% of students' actually rests on pooled response-level percentages, which limits the strength of the student-level claims. The descriptive findings are still useful for instructors and for future research on visual design in physics instruction, provided the claims are appropriately scoped.

major comments (3)

[Abstract; Section V.B] The abstract states that 'on average 40% of students found color to be helpful in matching and connecting related information or in separating and distinguishing distinct information,' but the data are percentages of pooled responses, not percentages of students. The surveys were anonymous and a single student could answer up to nine questions, so responses are non-independent; the paper itself acknowledges for the 'Color-coding' category that 'we cannot know if the "Color-coding" response population was the same for each question' (Section V.B). The same limitation applies to all aggregated percentages, including the headline figure. Because the abstract's number is the central quantitative takeaway, please rephrase to say 'responses' or, ideally, provide a student-level analysis (for example, counting each student only once per reason) and report both units of analysis explicitly.
[Section IV, Table III; Section V.C, Fig. 5] The Reason categories 'Matching/connecting info' and 'Separating/distinguishing' are not mutually exclusive; Section IV states that multiple Reason categories can apply to a single response if it makes distinct statements. The abstract's 'or' suggests a union of the two categories, but if the 40% figure is obtained by summing the separate category percentages, responses containing both reasons would be double-counted. Please specify how the 40% was calculated: is it the 'Specific reason subtotal' from Fig. 5 (the proportion of responses with one or more specific reasons), or the sum of two individual category percentages? If it is the latter, report the union or clearly state that the categories overlap.
[Title; Section I] The title and the stated purpose use causal language ('The effect of color-coding on students' perception'; 'whether the use of color-coding ... affected students' perceptions'), but the study design is a single-group descriptive survey with no control condition, no pre-post comparison, and no manipulation check. The data can support statements about students' reported perceptions and opinions, but not an estimate of an 'effect' of color-coding. Please temper the title and framing to describe a descriptive or exploratory study of students' perceptions, and explicitly acknowledge in the limitations that the design cannot separate the effect of color-coding from other course features or from students' general positivity toward the course.

minor comments (5)

[Section V (heading numbering)] The section heading 'V. Results and Discussion' is followed later by 'V. Conclusion'; the conclusion should be numbered as a new section (e.g., VI) to avoid duplicate section numbers.
[Figures 1 and 2] Figure 1 displays Scheme 2 while Figure 2 displays Scheme 1, which reverses the order in which the schemes are introduced in Section II; consider reordering the figures or their captions so that Scheme 1 appears first, matching the text.
[Table I, column headers] The column header 'Col 6' should be more descriptive, such as 'Schemes Used' or 'Scheme(s) in Use,' to match the content of the column.
[Section V.B] The claim that Question 8's high 'Color-coding' rate is not due to the word 'distinguish' in the prompt, because responses included details beyond the prompt, is plausible but speculative; please soften the assertion or provide supporting evidence that the details were not simply echoes of the prompt language.
[Appendix II.C] The inter-rater agreement is reported as raw pairwise agreement percentages; consider also reporting a chance-corrected agreement measure (such as Cohen's kappa or Fleiss' kappa) for the multi-rater coding, since raw agreement can be inflated by category base rates.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the empirical survey claims derive from independently coded open-ended responses, and the paper's caveats address the main validity concerns.

full rationale

The paper makes no formal derivation; its conclusions are descriptive summaries of anonymously collected open-ended survey responses. The vocabulary of the Reason categories ("Matching/connecting info", "Separating/distinguishing") overlaps with the intervention's motivating definitions, which is a surface resemblance, but the manuscript states that these categories "were not prescriptive, but rather emerged from evaluating survey responses for common themes" (Section IV), and all three authors coded independently with consensus. The headline "40% of students" figure is a pooled response-level percentage; the paper explicitly cautions that "we cannot know if the "Color-coding" response population was the same for each question" (Section V.B), which is a unit-of-analysis limitation rather than a circular reduction to the input. The only self-citation, Ref. 21 to the first author's under-review description of the color-coding strategies, is used to reference the designed materials; because the schemes are fully specified in the present paper and the conclusions rest on survey data, this citation is not load-bearing. No fitted parameter is renamed as a prediction, no uniqueness theorem is imported, and no known result is repackaged as an organizing framework.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The analysis has no fitted parameters or invented entities. All quantitative claims are derived from coded student responses, so the main burden rests on the coding categories and the self-report data, which are detailed in the paper.

assumptions (4)

domain assumption Student self-reports of perceived learning are treated as valid evidence of the intervention's helpfulness.
The study measures perception, not learning outcomes; the conclusion that color-coding is 'helpful to learning' relies on this assumption.
domain assumption Anonymous responses from the same students across multiple questions can be pooled and analyzed as independent observations.
Percentages are computed over responses, and without a student identifier the same student may contribute multiple responses, yet the abstract reports '40% of students'.
domain assumption The two cohorts (Fall 2020 and Spring 2021) experienced equivalent instruction and can be combined.
The paper states instructional materials were identical, but the cohorts differed in pandemic conditions and attendance modality, which could affect perceptions.
domain assumption Color-coding schemes were implemented consistently in slide-based videos and were never pointed out to students.
The study relies on students noticing the schemes without prompting; any inconsistency in implementation would affect the validity of the 'spontaneous noticing' claims.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The effect of color-coding on students' perception of learning in introductory mechanics." pith.science (2026). https://pith.science/paper/I5RV3QTW

@misc{pith2026241114605,
  author       = {Pith},
  title        = {Pith review of: The effect of color-coding on students' perception of learning in introductory mechanics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/I5RV3QTW}},
  note         = {Machine review of arXiv:2411.14605}
}

read the original abstract

We designed three color-coding schemes to identify related information across representations and to differentiate distinct information within a representation in slide-based instruction for calculus-based introductory mechanics. We found that students had generally favorable opinions on the use of color and that the few negative criticisms are easily addressed through minor modifications to implementation. Without having the color-coding schemes pointed out to them, a modest but consistent minority of students who found color helpful also described the color-coding schemes implemented, and about a quarter described the use of color in physics contexts as helpful even if they did not describe color-coding. We found that students particularly favored using color in mathematics and color-coding used to identify related variables, verbal definitions, and diagram elements. We additionally found that on average 40% of students found color to be helpful in matching and connecting related information or in separating and distinguishing distinct information, which were the motivating reasons for employing the color-coding schemes.

Figures

Figures reproduced from arXiv: 2411.14605 by the authors.

**Figure 4.** (color online) Percent of responses within each “Type” category, for all responses collectively (“All”) and for each question individually. The number in parenthesis after the cluster name is the number of responses to that question. Color-coding, color-context, color generic, other non-color, and off-topic are mutually exclusive with each other but not mutually exclusive with other categories, so percentages may no… view at source ↗

**Figure 5.** (color online) Percent of responses for each “Reason” category for all responses collectively (“All”) and for each question individually, by (top) broad category subtotals and (bottom) individual categories. The number in parenthesis after the cluster name is the number of responses for that question. “Specific Reason” and “Generic Reason” subtotals are the percent of responses which gave one or more reasons within … view at source ↗

**Figure 6.** (color online) Percent of responses for each “Reason” category among responses with a “Type” Category for which color is described as helpful, for all responses (“All”) and each question, by (left) subtotals and (right) individual categories. The number in parenthesis after the question number is the number of responses the specified “Type” category. “Specific Reason” and “Generic Reason” subtotals are the percent o… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 22 canonical work pages

[1]

Problem Solving and the Use of Math in Physics Courses

Edward F. Redish “Problem solving and the use of math in physics courses” presented at Conference World View on Physics Education in 2005: Focusing on Change, Delhi, August 21–26, 2005, accessed at http://arxiv.org/abs/physics/0608268

work page Pith review arXiv 2005
[2]

Brahmia Mathematization in introductory physics, Doctoral Dissertation, Rutgers University (20154)

Suzanne M. Brahmia Mathematization in introductory physics, Doctoral Dissertation, Rutgers University (20154). Retrieved from https://doi.org/doi:10.7282/T3FB51D8

work page doi:10.7282/t3fb51d8
[3]

Students’ difficulties with multiple representations in introductory mechanics

Dong-Hai Nguyen and N. Sanjay Rebello, “Students’ difficulties with multiple representations in introductory mechanics”, US-China Educ. Rev. 8 (5) 559-569 (2011)

work page 2011
[4]

Same to us, different to them: Numeric computation versus symbolic representation

Eugene Torigoe and Gary Gladding “Same to us, different to them: Numeric computation versus symbolic representation.” presented at Physics Education Research Conference 2006, Syracuse, New York. (2006) <https://www.compadre.org/Repository/document/ServeFile. cfm?ID=5263&DocID=2133>

work page 2006
[5]

Connecting symbolic difficulties with failure in physics

Eugene T. Torigoe and Gary E. Gladding, “Connecting symbolic difficulties with failure in physics” Am. J. Phys. 79 (1), 133–140 (2011)

work page 2011
[6]

Patterns of multiple representation use by experts and novices during physics problem solving

Patrick B. Kohl and Noah D. Finkelstein, “Patterns of multiple representation use by experts and novices during physics problem solving”, Phys. Rev. ST Phys. Educ. Res. 4 (1), 010111 (2008)

work page 2008
[7]

Differences in visual attention between those who correctly and incorrectly answer physics problems

Adrian M. Madsen, Adam M. Larson, Lester C. Loschky, and N. Sanjay Rebello “Differences in visual attention between those who correctly and incorrectly answer physics problems”, Phys. Rev. ST Phys. Educ. Res. 8 (1), 010122 (2012)

work page 2012
[8]

Student difficulties in connecting graphs and physics: Examples from kinematics

Lillian C. McDermott, Mark L. Rosenquist, Emily H. van Zee, “Student difficulties in connecting graphs and physics: Examples from kinematics”, Am. J. Phys. 55 (6) 503-513 (1987)

work page 1987

Show all 22 references

[9]

Multimedia and understanding: Expert and novice responses to different representations of chemical phenomena

For similar results in chemistry, see Robert. B. Kozma and Joel Russell, “Multimedia and understanding: Expert and novice responses to different representations of chemical phenomena”, J. Res. Sci. Teach. 34 (9), 949-968 (1997) 18

1997
[10]

Color as an Instructional Variable

Francis M. Dwyer, “Color as an Instructional Variable” AV Communication Review, 19 (4), 399-416 (1971)

1971
[11]

Effect of Color Coded Information on Students’ Levels of Field Dependence

David M. Moore and Francis M. Dwyer, “Effect of Color Coded Information on Students’ Levels of Field Dependence”, Perceptual and Motor Skills 72, 611-616 (1991)

1991
[12]

Relationship of Field Dependence and Color Coding to Female Students’ Achievement

David M. Moore and Francis M. Dwyer, “Relationship of Field Dependence and Color Coding to Female Students’ Achievement”, Perceptual and Motor Skills 93, 81-85 (2001)

2001
[13]

The Instructional Effect of Color in Immediate and Delayed Retention

Richard J. Lamberski, “The Instructional Effect of Color in Immediate and Delayed Retention”, in: Annual Meeting of the Association for Educational Communications and Technology, Research and Theory Division (Dallas, TX, May 1982)

1982
[14]

From Color Code to Color Cue: Remembering Graphic Information

Peggy A.P. Pruisner, “From Color Code to Color Cue: Remembering Graphic Information”, in: Visual Literacy in the Digital Age: Selected Readings from the Annual Conference of the International Visual Literacy Association (Rochester, NY, Oct 1993)

1993
[15]

The Influence of Color on Memory Performance: A Review

Mariam A. Dzulkifli, Muhammad F. Mustafar, “The Influence of Color on Memory Performance: A Review”, Malay J Med Sci 20 (2), 3-9 (2013)

2013
[16]

An eye-tracking study of how color coding affects multimedia learning

Erol Ozcelik, Turkan Karakus, Engin Kursun, and Kurat Cagiltay, “An eye-tracking study of how color coding affects multimedia learning”, Computers & Education 53, 445-453 (2009)

2009
[17]

The Effects of Highlighting on the math Computation Performance and Off-task Behavior of Students with Attention Problems

Suneeta Kercood and Janice A Grskovic, “The Effects of Highlighting on the math Computation Performance and Off-task Behavior of Students with Attention Problems”, Education and Treatment of Children, 32 (2), 231-241 (2009)

2009
[18]

Exploring the characteristics of an optimal design of digital materials for concept learning in mathematics: Multimedia learning and variation theory

Thomas. K. F. Chiu and Daniel Churchill, “Exploring the characteristics of an optimal design of digital materials for concept learning in mathematics: Multimedia learning and variation theory”, Computers & Education 82, 280-291 (2015)

2015
[19]

Thesis, Rowan University (2002)

Jennifer Valinote, The Effect of Color as a Visual Aid in Mathematics Instruction, M.S. Thesis, Rowan University (2002)

2002
[20]

Color Coding of Circuit Quantities in Introductory Circuit Analysis Instruction

Jana Reisslein, Amy M. Johnson, and Martin Reisslein, “Color Coding of Circuit Quantities in Introductory Circuit Analysis Instruction”, IEEE Transactions on Education, 58 (1), 7-14 (2015)

2015
[21]

Color-Coding Strategies for Multiple Representations

Brianna S. Dillon Thomas, “Color-Coding Strategies for Multiple Representations”, under review
[22]

Color-Coded Algebra

Bradley K. McCoy, “Color-Coded Algebra”, Phys. Teach. 59, 286–287 (2021) 19 Appendix I. Survey Administration This study was implemented in a calculus-based introductory mechanics course at a predominantly undergraduate state university (enrollment ~10k students) in the southe...

2021

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Problem Solving and the Use of Math in Physics Courses

Edward F. Redish “Problem solving and the use of math in physics courses” presented at Conference World View on Physics Education in 2005: Focusing on Change, Delhi, August 21–26, 2005, accessed at http://arxiv.org/abs/physics/0608268

work page Pith review arXiv 2005

[2] [2]

Brahmia Mathematization in introductory physics, Doctoral Dissertation, Rutgers University (20154)

Suzanne M. Brahmia Mathematization in introductory physics, Doctoral Dissertation, Rutgers University (20154). Retrieved from https://doi.org/doi:10.7282/T3FB51D8

work page doi:10.7282/t3fb51d8

[3] [3]

Students’ difficulties with multiple representations in introductory mechanics

Dong-Hai Nguyen and N. Sanjay Rebello, “Students’ difficulties with multiple representations in introductory mechanics”, US-China Educ. Rev. 8 (5) 559-569 (2011)

work page 2011

[4] [4]

Same to us, different to them: Numeric computation versus symbolic representation

Eugene Torigoe and Gary Gladding “Same to us, different to them: Numeric computation versus symbolic representation.” presented at Physics Education Research Conference 2006, Syracuse, New York. (2006) <https://www.compadre.org/Repository/document/ServeFile. cfm?ID=5263&DocID=2133>

work page 2006

[5] [5]

Connecting symbolic difficulties with failure in physics

Eugene T. Torigoe and Gary E. Gladding, “Connecting symbolic difficulties with failure in physics” Am. J. Phys. 79 (1), 133–140 (2011)

work page 2011

[6] [6]

Patterns of multiple representation use by experts and novices during physics problem solving

Patrick B. Kohl and Noah D. Finkelstein, “Patterns of multiple representation use by experts and novices during physics problem solving”, Phys. Rev. ST Phys. Educ. Res. 4 (1), 010111 (2008)

work page 2008

[7] [7]

Differences in visual attention between those who correctly and incorrectly answer physics problems

Adrian M. Madsen, Adam M. Larson, Lester C. Loschky, and N. Sanjay Rebello “Differences in visual attention between those who correctly and incorrectly answer physics problems”, Phys. Rev. ST Phys. Educ. Res. 8 (1), 010122 (2012)

work page 2012

[8] [8]

Student difficulties in connecting graphs and physics: Examples from kinematics

Lillian C. McDermott, Mark L. Rosenquist, Emily H. van Zee, “Student difficulties in connecting graphs and physics: Examples from kinematics”, Am. J. Phys. 55 (6) 503-513 (1987)

work page 1987