Address Biases in the Use of Language and Symbols¶

Strategy · All strategies
Evidence · 3 claims (3 for) · 4 studies (2 causal, 2 review), q2–q3 · 1 of 4 report an effect size · 2 claims rest on one study

Description¶

This strategy involves systematically auditing and revising the language, examples, names, images, and symbols used in learning materials, assessments, and classroom discourse so they are free from gender, racial, cultural, and ability-based stereotypes and respect the diverse identities of learners. It is carried out through bias review of materials, adoption of inclusive terminology (e.g., person-first or identity-first language as appropriate), and replacement of stereotyped example contexts with varied, representative ones.

Design Implications¶

Biased language and symbols impose extraneous processing and identity threat: stereotype threat research shows that cues signaling group-based expectations depress performance for targeted learners [~S], while inclusive representation supports belonging and persistence Self-efficacy predicts academic persistence. [+M]. Because attention is directed by salience, who appears in examples and who is cast in stereotyped roles shapes learners' implicit models of who "does" a subject Relevancy of emphasis directs attention. [+M].

Context¶

Requirements¶

• A bias review protocol applied to new and existing materials (names, pronouns, roles, imagery, idioms, units, currency, holidays)
• Current guidance on inclusive terminology for the learner population, updated periodically since norms evolve
• Diverse, non-stereotyped example contexts: varied names, professions, family structures, and ability representations across Case Studies, worked examples, and word problems
• Symbol and icon audit: flags, hand gestures, colors, and icons carry different meanings across cultures
• An audit process for existing materials: names used in examples, gendered defaults ("the engineer… he"), culturally narrow contexts, and imagery or symbols that signal a single demographic
• Replacement examples that represent diverse identities without tokenism — diversity in roles, contexts, and contributions, not just in decorative detail
• A norm-setting discussion with learners about inclusive language in discussion and peer feedback
• Ongoing review, since bias creeps back through unexamined defaults (Check-ins and learner feedback loops)

Constraints¶

• Superficial "diversity swaps" (changing a name without changing the surrounding context) can read as tokenism and fail to reduce stereotype threat [~M]
• Overcorrection that erases all cultural specificity produces bland, disengaging materials and removes the authenticity that supports Building Empathy [-W]
• Materials translated or localized without symbol and idiom review can introduce unintended meanings; literal translation alone is insufficient [-M]
• Audit effort competes with other design resources; without a checklist it is inconsistently applied [-W]
• Superficial or tokenistic representation can backfire: learners notice when diversity is cosmetic, and it can reinforce rather than reduce stereotype threat [~S]
• Learners may resist language changes they perceive as political, particularly when the rationale is not explained; framing changes around accuracy and audience reach reduces resistance
• Overly sanitized language can strip examples of the contextual richness that supports Analogical Reasoning and transfer; the goal is accurate, varied contexts, not context-free ones
• One-time audits decay — materials drift back to defaults without periodic re-review

Implementation Variability¶

• Light-touch: terminology and pronoun-neutral phrasing in new materials
• Structural: revising example role distributions across an entire curriculum (e.g., women and men appearing equally as scientists, caregivers, and authorities)
• Participatory: involving learners or community members in reviewing materials for bias
• Automated: linters and style guides (e.g., Microsoft's inclusive writing checker) flagging biased phrasing in authoring tools
• Materials-level: revising textbooks, problem sets, slide imagery, and interface icons
• Discourse-level: instructor modeling and gentle correction of biased language in live discussion
• Assessment-level: ensuring scenario-based items do not require cultural background knowledge unrelated to the construct being measured
• Learner-facing: engaging students in auditing and rewriting biased materials themselves, which doubles as critical literacy practice

Target Learners¶

• Learners from groups historically underrepresented or stereotyped in a subject domain (e.g., women in physics, Black students in mathematics), for whom identity-relevant cues affect performance and belonging [~S]
• Multilingual learners, for whom idioms, culture-bound references, and ambiguous symbols add decoding load beyond the target content — favor Accessible Vocabulary & Syntax [+M]
• Learners with disabilities, who benefit from person-centered rather than deficit-framing language [+W]
• Learners from marginalized or underrepresented groups in a domain, for whom cue removal reduces identity threat and supports belonging [~S]
• All learners benefit from clearer, more accurate materials; stereotype threat effects are strongest for learners who are highly identified with the domain [~S]
• Belonging interventions are especially effective at transition points (first year, first course in a discipline) where belonging uncertainty is highest [~S]

Target Learning Goals¶

• All content goals: bias reduction is a cross-cutting materials-quality strategy, not a content goal itself
• Equity of access: removing identity-based barriers so content goals can be pursued on equal footing
• Social-emotional goals: modeling respectful language supports classroom norms and Building Empathy
• Affective and climate goals: fostering belonging, respect, and willingness to participate
• Critical literacy goals: when learners audit materials themselves, they build skills in analyzing how language constructs social meaning

Instructions¶

1. Establish a bias review checklist covering names, roles, pronouns, imagery, idioms, symbols, and cultural assumptions.
2. Audit existing high-use materials, starting with assessments and Advance Organizers that frame a topic.
3. Revise examples so that diverse identities appear across a full range of roles and statuses, not only in token positions.
4. Replace culture-bound idioms and ambiguous symbols with plain language and universally interpretable icons, consistent with Clear Structure & Presentation.
5. Re-review periodically and after major terminology shifts; treat bias review as maintenance, not a one-time fix.

Claims¶

• Single-person images are significantly biased toward light-skinned individuals, while multi-person images show more balanced gender and skin-tone diversity [+W]

Related Strategies¶

• Accessible Syntax — sentence-level clarity work that pairs naturally with bias review, since both reduce unnecessary processing load
• Activating Prior Knowledge — biased materials can fail to activate, or actively contradict, the prior knowledge of diverse learners
• Building Empathy — auditing materials for bias is itself an empathy-building exercise for designers and learners
• Intentionally activate symbols so they assist transformative learning
• Use cultural informants from the target community to buffer against stereotyping and bias

Examples¶

• Illustrative Mathematics — K–12 curriculum whose problem-writing guidelines require varied, non-stereotyped names and contexts in word problems.
• Microsoft inclusive writing guidance — built into Microsoft Editor, flags gendered and ableist phrasing in documents in real time.
• WCAG 2.1 (W3C) — while focused on accessibility, its guidance on meaningful symbols, icons, and non-text content overlaps with symbol-bias review for materials design.
• Physics problem-set revision — Replacing a problem set in which all "engineers" and "scientists" were male and all "assistants" female with varied role–identity pairings; studies of classroom cues show such representation affects women's engagement in STEM settings [~S].
• CS curriculum imagery — Redesigning computer science classroom decor and course websites so that objects and imagery do not signal a single (male, tech-geek) culture; ambient cues measurably affect women's interest in entering computing environments [~S].
• Belonging-framing at course start — Pairing inclusive materials with a brief message that identity-based doubts are common and temporary; such interventions improved academic outcomes for marginalized students in a randomized field study [~S].
• Learner-led audits — Having students in a technical writing course audit a company's user documentation for biased language and propose revisions, turning the strategy into authentic practice with a real audience.

Key Sources¶

• Steele, C. M. (1997). A threat in the air: How stereotypes shape intellectual identity and performance. American Psychologist, 52(6), 613–629. doi:10.1037/0003-066X.52.6.613
• Walton, G. M., & Cohen, G. L. (2011). A brief social-belonging intervention improves academic and health outcomes of minority students. Science, 331(6023), 1447–1451. doi:10.1126/science.1198364
• Good, C., Aronson, J., & Inzlicht, M. (2003). Improving adolescents' standardized test performance: An intervention to reduce the effects of stereotype threat. Journal of Applied Developmental Psychology, 24(6), 645–662. doi:10.1016/j.appdev.2003.09.002
• Gay, G. (2018). Culturally Responsive Teaching: Theory, Research, and Practice (3rd ed.). Teachers College Press.
• Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69(5), 797–811. doi:10.1037/0022-3514.69.5.797
• Cheryan, S., Plaut, V. C., Davies, P. G., & Steele, C. M. (2009). Ambient belonging: How stereotypical cues impact gender participation in computer science. Journal of Personality and Social Psychology, 97(6), 1045–1060. doi:10.1037/a0016239
• Murphy, M. C., Steele, C. M., & Gross, J. J. (2007). Signaling threat: How situational cues affect women in math, science, and engineering settings. Psychological Science, 18(10), 879–885. doi:10.1111/j.1467-9280.2007.01995.x
• Gay, G. (2010). Culturally Responsive Teaching: Theory, Research, and Practice (2nd ed.). Teachers College Press.