This dataset contains the coded lexical and contextual data used in a corpus-informed analysis of emotion-related vocabulary in Italian as a foreign language (IFL) textbooks at the beginner level (CEFR A1) used in Polish lower secondary education. The dataset is based on four textbooks from two series: Progetto Italiano Junior (Marin, 2017; 2018) and Va bene! (Kaliska & Kostecka-Szewc, 2021). All materials were analysed in their printed form. The dataset includes all lexical items identified as emotion-related based on their presence in the ANEW-IT database (Montefinese et al., 2014), which provides normative ratings of valence and arousal for Italian words. Each entry in the dataset corresponds to a single token occurrence of an emotion-related lexical item. The dataset includes both surface forms as they appear in the textbooks and their corresponding base forms (lemmas) as listed in ANEW-IT. For each token, the dataset provides contextual, linguistic, and affective information. The variables included are: • textbook and series identification • unit/chapter, page number, and exercise reference • material type (e.g., dialogue, reading text, exercise, review) • pedagogical focus (e.g., grammar, vocabulary, comprehension, mixed) • word form (surface form) and lemma • part of speech (POS) • contextual sentence or description of occurrence • frequency measures (FreqColfis, Ln_Colfis) • affective ratings (valence and arousal, scale 1–9) The dataset enables replication of the quantitative analyses reported in the study, including token counts, type–token ratios, valence and arousal distributions, and comparisons across textbook series and pedagogical contexts.