Jockers/Rinker sentiment lexicon — Jockers

The Jockers/Rinker hand-curated sentiment lexicon (originally distributed as lexicon::hash_sentiment_jockers_rinker, MIT-licensed). Values are continuous in [-1, 1]. Of the 11,710 entries in the upstream source, 13 fell outside [-1, 1] (e.g., overly = -2, could have = -1.05, too much = -2). These are shifter / amplifier / counterfactual coefficients used by sentimentr's sliding-window aggregator to override neighboring sentiment, not independently sentiment-bearing words; they have been dropped here to avoid artificially biasing per-text means. Used by score_valence(include_jockers = TRUE).

Usage

Jockers_norms

Format

A data frame with 11,697 rows and 2 columns:

word: lowercased term (character)
valence: sentiment in [-1, 1] (numeric)

Source

Originally compiled by Matthew Jockers for narrative/literary sentiment analysis (syuzhet package); the variant used here is the Rinker-extended version distributed via the lexicon R package. Note: validated primarily on narrative/fiction text rather than person-perception research specifically. See data-raw/build_warriner_jockers.R for the build recipe.