Reasons and Challenges of Adopting Flexitarian Diet: A Mixed-Methods Approach Using Natural Language Processing

Abstract

In this study, we analyzed data from a large sample (n = 778) of individuals who identified as flexitarians, using Brazil as a case study. We applied Natural Language Processing (NLP) to characterize the reasons why individuals self-identify as flexitarians. We explored how these self-perceptions contribute to the construction of a flexitarian identity and identified the challenges this group faces in adopting a flexitarian diet. Our results indicate that there is no homogeneous flexitarian identity, but rather a diversity of categories, organized into five distinct clusters: ethical flexitarians (those who express greater concern for the environment and animals), flexible vegetarians (those who often express the desire to become vegetarian but still consume meat), meat lovers (those who enjoy meat but do not feel compelled to consume it), moderate consumers (those who refrain from meat on certain days of the week), and meat reducers (those who actively reduce their meat consumption). The latter three groups are more similar, as they tend to reduce their meat intake due to the belief that meat is not essential in their diet, although they still enjoy its taste. Additionally, we found that the main challenge faced by our participants is directly related to the social environment, specifically the difficulty in finding alternatives to meat, which makes the love for meats taste and meat-centric eating habits secondary challenges.