Analyzing the Representation of Body Image in Women’s Magazines Through NLP

Abstract

In the age of media and widespread digital presence, there is increasing pressure to conform to unrealistic beauty standards. The media places significant emphasis on body image and appearance-related features, with recent emphasis on the glamorization of a thin body ideal. The idolization of unrealistic body types can have dangerous psychological and physical impacts on women, such as body dysmorphia and disordered eating. This computational study compares and analyzes how body image is portrayed in four online women’s magazines — Woman’s World, Seventeen, Shape, and Cosmopolitan. In this study, a dataset totaling over 1000 articles was sourced from the online publications of the four magazines. Studying how women’s magazines portray body image can be useful in understanding the quantitative measures and raising awareness of the effects of such language. Through the application of Natural Language Processing (NLP) methods of nltk relative frequency tasks, we aim to interpret the appearance of body-related language pertaining to thinness, weight loss, and body image. Our results show consistent usage of body-image language for all four magazines. Out of the four magazines, Woman’s World has the most significant frequency of language related to weight loss, thinness, and body composition. Both Shape and Woman’s World also have prevalent language related to changing body shape.