Semantic Feature Analysis¶

Strategy · All strategies
Evidence · 2 claims (2 for) · 6 studies (4 causal, 2 quant-synthesis), q3–q4 · 2 of 6 report an effect size

Description¶

Semantic Feature Analysis (SFA) is a vocabulary and comprehension strategy in which learners complete a grid: related concepts (or words) form the rows, and features or properties form the columns. Learners mark each cell (+/−, or a continuum) to indicate whether the feature applies, then discuss and justify their markings. The grid makes the semantic relationships among a word set explicit and forces learners to attend to fine-grained distinctions between similar concepts.

Design Implications¶

SFA works because it requires learners to compare and contrast related concepts systematically, supporting abstraction of shared and distinguishing features Multiple contrasting cases support abstraction. [+M]. Completing the grid also activates prior knowledge before new content is encountered Activation improves learning. [+M]. The discussion phase — where learners defend contested cells — is where much of the learning occurs; grids completed silently and never discussed produce weaker gains.

Context¶

Requirements¶

• A carefully chosen set of related concepts (typically 5–10) that share a category but differ on meaningful dimensions
• Features that discriminate among the concepts, not generic attributes
• Teacher modeling of the analysis process before independent or group completion
• A follow-up discussion or writing task that uses the grid's distinctions

Constraints¶

• Ineffective when features are trivial or the word set is unrelated — the grid then becomes rote cell-filling with no relational processing [-M]
• Poorly chosen feature lists can introduce misconceptions when binary +/− markings flatten genuinely continuous or fuzzy distinctions [~M]
• Time-intensive relative to vocabulary gains; the overhead is hard to justify for words that do not warrant deep relational analysis [-W]
• Less effective for learners with very low prior knowledge of the domain, who lack the knowledge to activate and may mark cells arbitrarily [~M]

Implementation Variability¶

• Prediction-first: learners complete the grid before reading, then revise after reading — turning the grid into a comprehension monitor
• Continuum markings: replace +/− with ratings (e.g., "always/sometimes/never") for fuzzy features
• Student-generated features: learners propose the columns themselves, deepening analysis at the cost of time
• Digital grids: collaborative spreadsheets allow simultaneous completion and visible disagreement, which fuels discussion

Target Learners¶

• Elementary and adolescent readers building domain vocabulary (science, social studies) Multiple contrasting cases support abstraction. [+M]
• English language learners, for whom the grid makes connotation and collocational distinctions visible [~M]
• Students with learning disabilities benefit when the grid is teacher-constructed and analysis is modeled [~W]

Target Learning Goals¶

• Relational vocabulary knowledge: knowing how words are similar to and different from one another, not just definitions
• Categorization and concept formation: identifying defining versus non-defining features
• Comprehension monitoring: using the grid to check and revise understanding of a text

Instructions¶

1. Select a category and 5–10 related concepts; list them as rows of a grid.
2. Generate 5–8 discriminating features as columns (model this step aloud — see Analogies for connecting new terms to known ones).
3. Model completing two or three rows, thinking aloud about why a feature does or does not apply.
4. Have learners complete the remaining cells in pairs or small groups, requiring justification for contested cells.
5. Lead a whole-class discussion of disagreements; revise cells based on evidence from text or experience.
6. Follow with a writing or discussion task that requires using the distinctions the grid surfaced.

Related Strategies¶

• Frayer Model — single-concept deep analysis; SFA extends the same feature-analysis logic across a whole word set
• Activating Prior Knowledge — SFA is a structured vehicle for this broader move
• Word Walls — SFA grids can anchor a wall display of related terms

Examples¶

• Science: a grid of vertebrate classes (mammal, bird, reptile, amphibian, fish) against features (warm-blooded, live birth, scales, aquatic) — contested cells drive the lesson
• Social studies: forms of government (democracy, monarchy, oligarchy, theocracy) against features (who holds power, how leaders are chosen, citizen rights)
• Reading instruction: the strategy is a core component of published vocabulary programs such as Word Wisdom and appears in the International Reading Association's strategy collections (Reading Rockets: Semantic Feature Analysis)

Key Sources¶

• Schwartz, R. M., & Raphael, T. E. (1985). Concept of definition: A key to improving students' vocabulary. The Reading Teacher, 39(2), 198–205.
• Bos, C. S., & Anders, P. L. (1990). Effects of interactive vocabulary instruction on the vocabulary learning and reading comprehension of junior-high learning-disabled students. Learning Disability Quarterly, 13(1), 31–42. doi:10.2307/1510390
• Anders, P. L., & Bos, C. S. (1986). Semantic feature analysis: An interactive strategy for vocabulary development and text comprehension. Journal of Reading, 29(7), 610–616.
• Nagy, W. E. (1988). Teaching vocabulary to improve reading comprehension. International Reading Association.