Research indicates that beliefs on the locus of the primary source of learning can have negative effects on learning behaviors in school (Mercer & Ryan, 2011). To an increasing extent, young people in Sweden acquire English outside school through different cultural practices (Sundqvist & Sylvén, 2012). At the same time, students lack enthusiasm in English lessons, which can lead to a reduced investment of effort in the language classroom (Henry, 2014). However, learning a language requires both interest and effort. In psychological research, the capacity of showing consistency in interest and perseverance to achieve a goal, such as learning a language, is defined as grit (Duckworth et al., 2007). Studies investigating the association between grit and language achievement remain scarce. The present study contributes to filling this gap by investigating whether the effects of grit differ across different performance domains (i.e., English, Swedish and Mathematics). An additional purpose is to investigate whether there are differential effects of grit for subgroups of students with different SES, achievement levels, and gender. A total of 4646 compulsory school students born in 1992 were extracted from the Evaluation Through Follow-up Database (ETF). Multiple-group path analysis was used to examine the mechanisms of the grit facets (i.e., interest and perseverance) across time and subgroups. While both interest and perseverance predicted grades in Swedish and Mathematics, interest alone predicted grades in English. The low SES high achieving group also showed higher consistency in interest and perseverance than the other groups, indicating that grit could be more valuable in face of adversities.