The Complementarity of STEM Skills and English Fluency: Female Representation in College STEM in India

• In Event: Digital Transitions in Education: Resilience, Gender Dynamics, and Technological Impacts

Proposal

1. How is the topic relevant to CIES 2025?

In the era of artificial intelligence (AI) that affects every aspect of our lives, equitable and inclusive STEM workforce development is a pressing need for higher education. STEM degrees are required in most jobs in emerging fields. However, there is a significant under-representation of many individuals in STEM fields, especially women. Given that the current AI workforce broadly consists of one homogeneous group, their world views, perspectives, and preferences are likely to be calcified into AI models and systems. Hence, STEM workforce development, including college-level STEM education, should be explored in a way to achieve STEM diversity “so that everyone’s story is part of the models we are building” (Boinodiris & Rudden, 2023, p.25). Using a comparative perspective, this study explored determinants of college STEM diversity in gender, focusing on female youth in India as an under-represented sub-group in STEM.

2. What framework guided the research questions?

This study adopted an interdisciplinary conceptual framework integrating human capital theory, social reproduction theory, and intersectionality to understand the context within which the following determinants interact:

3. How are sources of information used to inform choices about data collection and analysis?

Much of the current literature understand female under-representation in college STEM as a result of individual traits, e.g., STEM self-efficacy – subjective judgement of one’s ability to complete a certain task in STEM. However, the recent international comparative studies using the PISA data statistically have demonstrated that girls and women are under-represented in STEM fields in gender-egalitarian, high-income countries such as the U.S. In contrast, their counterparts in middle-income countries tend to be represented in STEM fields while they may tend to be more affected by sociocultural norms. This evidence suggests that there may be some non-individual, country-specific characteristics that would amplify or attenuate the effects of biological sex and gender on female students’ choice of college major in respective countries.

Nevertheless, these studies do not sufficiently explore why and how this phenomenon occurs in middle-income countries while emphasizing only one factor as an enabler - such as the lack of cultural gender beliefs that STEM fields are “masculine.” This is because they were not necessarily conducted by researchers of international comparative education while each academic discipline tends to look at only one factor that is in fact intertwined intricately with other various factors. That is, recent qualitative research by Indian and Indian-origin researchers suggests that for some groups in India, STEM careers are highly desirable, despite conservative social attitudes toward women. Their findings show that female representation in STEM fields at college may be a social product of various factors and systems - income, gender, school type, geography, and foreign language ability under a culturally enabling but unequally meritocratic environment for female students to choose a STEM major at college. Hence, a deeper, contextual understanding of female representation in STEM fields at college is lacking in the current international comparative studies on this research problem.

Hence, the study attempts to answer the research question: How does parental socioeconomic status, mediated by English language proficiency and moderated by gender, school type, and geographical location, relate to a student’s college major in India?

4. How do the research methods and results support the conclusions drawn from the data?

Methodologically, it applied three-level, generalized structural equation modeling (MSEM) with the India Human Development Survey (IHDS) II 2011-2012 dataset (N =11,210). With student choice of major as the binary dependent variable, the present study used three-level, generalized MSEM (level 1: students, level 2: districts, level 3: states) to statistically investigate the effects of parental socioeconomic status (SES), gender, students’ English language proficiency, school type, and geography on a student’s college major as a combination of (a) simple mediation (b) single-level moderated mediation and (c) cross-level interactions for statistical contextual understanding.

Results suggest that female representation in STEM fields in India is conditional on English fluency due to the English-STEM complementarity embedded in Indian universities and hiring systems. Students with high-level English were 2.38 times more likely to choose a college STEM major than their counterparts with less English skills. When girls were fluent in English, they were 5.24 times more likely to choose STEM majors than girls who were not. Specifically, girls who were fluent in English were significantly more likely to choose a STEM major than boys. To develop such high-level English skills, Indian students are likely to require English-medium K-12 schooling provided by private schools. Children who studied at English-medium schools were 9.17 times more likely to develop high English proficiency as compared to those who studied at non-English medium, public schools. Given that parents are responsible for the development of their children’s English skills, children’s English fluency is likely to depend on family, access to English-medium learning environments, and parents’ choice of such schools. As a result, the most advantaged group of students (i.e., male students from English-speaking states who studied at English-medium schools) were 4.39 times more likely to choose a college STEM major than the most disadvantaged group of students (i.e., female students from non-English-speaking states who studied at non-English-medium schools). The convergence diagnostics indicated successful convergence (the final PSR = 1.02), validating a successful model-fit. In summary, STEM-English complementarity and parental role in development of children’s English skills appear to explain the mechanism of an Indian student’s choice of college STEM majors.

5. How original is the contribution? What do we learn that we did not know, and why is it important?

This study empirically demonstrates that female representation in STEM can derive from a new and complex form of social disadvantage and inequality by linking STEM skills with English fluency that is functioning as cultural capital as well as human capital. Therefore, it suggests that advocating positive gender beliefs about female STEM participation alone may unintentionally strengthen other forms of educational inequality among girls and women. Hence, the results point to how we might envision STEM education in a digital society where diversity, equity, and inclusion similarly matter.

Author

• Kazumi Homma, The George Washington University