Sexist Textbooks: CGD Working Paper 678

FOCUS

In their 2024 working paper "Sexist Textbooks," Lee Crawfurd, Christelle Saintis-Miller and Rory Todd, researchers with the Center for Global Development, Washignton D.C., investigate gender bias in textbooks from 34 countries. The paper explores whether this bias correlates with the legal and economic status of women in these countries. It analyses 1,255 textbooks using word lists of gendered words, family and work terms.

To assess gender stereotypes, the paper measures the co-occurrence of gendered names and pronouns with specific terms (for example ‘he’ and ‘doctor’). It also uses word embeddings to analyse stereotypes and biases and applies part-of-speech tagging to reveal a skewed distribution of certain adjectives and verbs across genders. This reveals patterns regarding the depiction of male and female characters and their actions in textbooks.

The study uncovers significant gender bias in the textbooks, with stereotypes more prevalent in lower and lower-middle-income countries. However, the authors acknowledge limitations, such as relying solely on English-language textbooks, potential imprecision in methods, and difficulties distinguishing content to be read by students and other text such as information about the publisher. They suggest further research into gender representation in textbook images, which was beyond the scope of this study.

    FACTOIDS

  1. The paper analyses 1,225 textbooks for Classes 4 to 13 from 34 countries including Afghanistan, Australia, Bangladesh, Belize, Bhutan, Dominica, Ethiopia, Guyana, India, Jamaica, Kenya, Kiribati, Lesotho, Liberia, Malawi, Maldives, Namibia, Nigeria, Pakistan, Papua New Guinea, Rwanda, Samoa, Sierra Leone, Solomon Islands, South Africa, South Sudan, Sri Lanka, St. Kitts and Nevis, Tonga, Uganda, the United Kingdom, the United States, Zambia and Zimbabwe.

  2. Across the full sample, the study finds that ‘male words’ occur more than twice as frequently (178,142) as ‘female words’ (82,113). The lowest representation of women and girls in textbooks is found in Afghanistan, Pakistan, Sri Lanka and South Sudan, where fewer than one-third of gendered words are female. While subjects like math and science show extremely skewed representation in favour of men, humanities and religious studies show the lowest female representation. The paper states that home economics is the subject wherein female representation is the highest.

  3. Amongst the twenty most frequently mentioned professions all except ‘nurse’ are associated more with male terms than with female terms. According to the data, occupations like ‘nurse’ and ‘housekeeper’ co-occur with female words most frequently, whereas those like ‘blacksmith’ and ‘physicist’ co-occur most with male terminologies.

  4. While female words are more likely to be associated with domains of home, family or appearance, male words seem to be associated with work or achievements. Strong male bias is seen in words like ‘leader’, ‘powerful’, and ‘authority’ whereas words like ‘slim’ and ‘wedding’ have a strong female bias.

  5. The study utilizes several indices, including the Social Institutions and Gender Index (SIGI), the Women, Business and the Law Index (WBL), the Girl’s Education Policy Index (GEPI), the World Bank Gender Parity Index in Gross Secondary Enrolment, as well as data on teenage marriage and female members of parliament from the OECD, to analyse gender representation and its correlation with broader gender equality measures.

  6. The study finds a strong correlation between gender representation in textbooks and national gender bias measures. A standard deviation rise of 1 point in the SIGI, WBL, or GEPI indices increases the share of female words in textbooks by 5-6 percentage points. A similar rise in gender parity in secondary education or female members of parliament increases the share by two percentage points. Likewise, higher rates of girl marriage are linked to fewer female words in textbooks.

  7. The authors hope to identify a new focus for education policy agendas. They advocate for prioritising gender representation in future textbook revisions and suggest that their model of assessing representation can be easily replicated with existing texts as well as successive editions to pre-emptively examine drafts for gender bias.


    Focus and Factoids by Adrita Bhattacharya.

AUTHOR

Lee Crawfurd, Christelle Saintis-Miller and Rory Todd

COPYRIGHT

Center for Global Development

PUBLICATION DATE

Feb, 2024

TAGS

#women #education #schools #gender-inequality

SHARE