The study presents a comprehensive set of emotional norms for 13,915 English lemmas, expanding on the previously used ANEW norms for 1,034 words. The new dataset includes ratings for valence (pleasantness), arousal (intensity of emotion), and dominance (degree of control), and covers a wide range of semantic categories, including diseases, occupations, and taboo words. The data were collected through crowdsourcing, involving over 1,827 participants who rated words on a 9-point scale across three dimensions. The dataset includes demographic information such as age, gender, and education level, allowing for analysis of differences in emotional ratings across these variables. The study found that valence and dominance ratings were more consistent across participants, while arousal ratings showed greater variability. Correlations between the dimensions showed a U-shaped relationship between arousal and valence, and between arousal and dominance. Valence and dominance were found to be linearly related, with higher valence associated with higher dominance. The study also found that gender differences influenced emotional ratings, with females showing more extreme ratings for certain words compared to males. The dataset provides a rich source of information for researchers studying emotions, semantics, and text-based sentiment analysis. It allows for the automatic estimation of emotional values of new words and supports the analysis of semantic memory and language processing. The study highlights the importance of considering demographic factors and the variability in emotional ratings when interpreting data. The dataset is available for use in future research and has the potential to enhance the accuracy of computational models in estimating emotional states and attitudes.The study presents a comprehensive set of emotional norms for 13,915 English lemmas, expanding on the previously used ANEW norms for 1,034 words. The new dataset includes ratings for valence (pleasantness), arousal (intensity of emotion), and dominance (degree of control), and covers a wide range of semantic categories, including diseases, occupations, and taboo words. The data were collected through crowdsourcing, involving over 1,827 participants who rated words on a 9-point scale across three dimensions. The dataset includes demographic information such as age, gender, and education level, allowing for analysis of differences in emotional ratings across these variables. The study found that valence and dominance ratings were more consistent across participants, while arousal ratings showed greater variability. Correlations between the dimensions showed a U-shaped relationship between arousal and valence, and between arousal and dominance. Valence and dominance were found to be linearly related, with higher valence associated with higher dominance. The study also found that gender differences influenced emotional ratings, with females showing more extreme ratings for certain words compared to males. The dataset provides a rich source of information for researchers studying emotions, semantics, and text-based sentiment analysis. It allows for the automatic estimation of emotional values of new words and supports the analysis of semantic memory and language processing. The study highlights the importance of considering demographic factors and the variability in emotional ratings when interpreting data. The dataset is available for use in future research and has the potential to enhance the accuracy of computational models in estimating emotional states and attitudes.