A Durability and Cross-Language Transfer Benchmark for a Validated Teaching-Feedback Classification Protocol
2026-07-13 • Computation and Language
Computation and LanguageMachine Learning
AI summaryⓘ
The authors studied a way to automatically classify student feedback comments by topic and sentiment. They tested if this method still works well with newer language models and if it can be used for English as well as Spanish. They found that while the newest models help with topic classification in Spanish, simpler models work just as well for detecting sentiment and for English data. This means the choice of model depends more on practical needs than on the method itself.
teaching evaluationsentiment analysisthematic classificationtransformer embeddingslarge language modelscross-validationintra-annotator reliabilitySpanish corpusEnglish corpus
Authors
Esteban U. Vega Barajas
Abstract
Institutions collect far more open-ended teaching-evaluation feedback than they read. A prior study introduced a validated protocol for classifying such comments by thematic category and sentiment, built from a documented annotation guide, an intra-annotator reliability measurement, stratified cross-validation, and a held-out evaluation on a Spanish institutional corpus with a frozen-encoder design. Two questions limit its reuse: whether a protocol fixed to 2019-era frozen embeddings stays competitive as representation methods advance, and whether it transfers to a second language. We re-run it on the original Spanish data across three representation generations, sparse lexical features, frozen transformer embeddings, and prompted large language models, and transfer its sentiment task to English with a balanced 45,000-comment corpus checked against an aspect-labeled education dataset. Treating paired comparisons as descriptive, we find the protocol durable: a 2026 frontier model posts the highest thematic F1 on the hardest Spanish task, yet shows no sentiment advantage over a cheap model and no descriptive separation from it on English, so model choice is a deployment decision, not a property of the method.