Teenage girls routinely believe they are better at English than maths - and boys believe the opposite - even when there is no difference in actual ability, research suggests.

The US study also suggests that children may develop such beliefs by the end of elementary school (typically attended between from age 4-14), when some students are already describing themselves as “math people”.

As predicted by the academics, males generally had higher “math self-concept” than females, and females had higher English self-concept than males. The finding echo previous research that suggested teachers also underrate girls’ ability in maths.

Students may perceive themselves “competent in multiple domains early in adolescence”, say University of California researchers Osman Umarji and Jacquelynne Eccles, but “over time, they gravitate towards seeing themselves as more of a ‘math’ or ‘verbal’ person” - with the risk of dismissing careers in which they might excel, such as engineering and technology.

They give several potential explanations for the disparity in outlook between the genders, including:

• Boys and their parents often attribute their success to talent, while girls and their parents often see effort as the main factor, which over time may lead to gender differences in maths “self-concept”.
• Males brag about their maths competence more than women, leading to girls making “downward social comparisons about their ability” relative to boys in their class.
• “Gender-role stereotyped socialisation” sees females place higher value on making occupational sacrifices for one’s family, while males place more value on fame, earning a more money, seeking out more challenging tasks and doing work that involves maths and computers.
• The perception of maths-intensive careers such as computer science, engineering, and physics as “nerdy” - stereotypes that may be incompatible with gender stereotypes of women.

The research - The Development of Adolescents’ Math and English Self-Concept Patterns and their Associations with College Major Selection - is being presented today at the American Educational Research Association’s annual meeting in San Antonio, Texas.

The data used in the research come from the Michigan Study of Adolescent and Adult Life Transitions (MSALT), a longitudinal study that began in 1983 and spans sixth grade (around 11 years old) to three years after high school (around 21 years old). The researchers analysed a subsample of 804 students from middle-class districts who reported being in college at age 21.

Want to keep up with the latest education news and opinion? Follow Tes on Twitter and like Tes on Facebook