Semantic Factor Analysis of Language Model Embeddings

---

Documentation for package ‘semanticfa’ version 0.5.0

• DESCRIPTION file.
• User guides, package vignettes and other documentation.
• Package NEWS.

Help Pages

semanticfa-package	semanticfa: Semantic Factor Analysis of Language Model Embeddings
as_psych	Coerce to psych fa Object
as_psych.sfa	Coerce to psych fa Object
big5	IPIP Big Five 50-Item Inventory with Qwen3 Embeddings
plot.sfa_coverage	Plot a Content-Validity Audit
plot.sfa_coverage_battery	Plot One Factor of a Content-Validity Audit Battery
semanticfa	semanticfa: Semantic Factor Analysis of Language Model Embeddings
sfa	Semantic Factor Analysis
sfa_anchor	Construct-Label and Centroid Anchoring
sfa_build_bank	Pre-Embed a Set of Texts into a Reusable Bank
sfa_build_region	Build a Construct Region from a Text Corpus
sfa_build_regions	Build Many Construct Regions in One Corpus Pass
sfa_cd	Comparison-Data Misfit Profile
sfa_clear_cache	Clear Embedding Cache
sfa_combine_banks	Combine Embedding Lookups
sfa_congruence	Compare Semantic and Empirical Factor Structures
sfa_corplot	Heatmap of an Item-by-Item Similarity Matrix
sfa_coverage	Audit the Content Validity of a Scale Against a Construct Region
sfa_cross_matrix	Extract the Numeric Matrix from a Cross-Audit
sfa_deletion_gaps	Newly Uncovered Content After an Item-Set Change
sfa_dimselect	Embedding-Dimension Selection by EGA Depth Optimization
sfa_ekc	Empirical Kaiser Criterion for Embedding Similarity Matrices
sfa_embed	Embed Item Text with a Language Model
sfa_embedding_bank	Use a Pre-Generated Embedding Bank as an Embedder
sfa_gaps	Gap Table from a Coverage Audit
sfa_install_python	Provision the Python Environment for Embedding
sfa_itemplot	2-D Item Map (t-SNE, UMAP, PCA, or MDS)
sfa_item_fit	Vet a Candidate Scale Item Before Data Collection
sfa_jinglejangle	Detect Jingle and Jangle Fallacies Across Scales
sfa_leximax	Leximax: Rotate Factor Axes Toward the Construct Lexicon
sfa_lexmap	Build the Lexical Map for Leximax Rotation
sfa_load_npz	Load Pre-generated Embeddings from a NumPy .npz File
sfa_load_region	Load a Saved Construct Region
sfa_map	Velicer's Minimum Average Partial for Similarity Matrices
sfa_name	Name the Factors of an sfa Fit
sfa_nameability	Nameability of an Orientation
sfa_naming_instruction	The Default Naming Instruction
sfa_nfactors	Unified Factor Retention Diagnostics
sfa_nli_matrix	Signed Item Similarity from Natural Language Inference
sfa_parallel	Embedding-Adapted Parallel Analysis
sfa_pool	Fetch or Build the Candidate Pool for a Naming Model
sfa_project	Semantic Projection onto Bipolar Axes
sfa_redundancy	Detect Redundant (Near-Duplicate) Items
sfa_reembed_region	Re-Embed a Construct Region Under Another Encoder
sfa_region_bank	Use a Region's Own Stored Embeddings as an Embedder
sfa_semk	Calibrated Semantic Factor Retention (sem-k)
sfa_similarity	Compute Embedding Similarity Matrix
sfa_simplify	Response-Free Scale Simplification
sfa_tsneplot	2-D Item Map (t-SNE, UMAP, PCA, or MDS)