Students who feel confident in English often get good grades. But which comes first, confidence or good grades? Research says both. Good grades build confidence, and confidence helps later grades a little.

In a 2007 study of 303 American university students learning French, one kind of confidence was most closely linked to grades. It was the belief "I can manage my own study." So write down small successes every day.

Students who are confident in English tend to get better grades, but does confidence cause success, or does success cause confidence? According to a 2011 review by Marsh and Martin, the answer is both. Past grades raise students' views of their own ability, and those views also raise later grades.

However, only beliefs about a specific subject mattered, while general self-esteem had little effect. A 2004 analysis of long-term studies found that the effect of self-beliefs on later achievement was small but positive. Which kind of confidence matters most in language learning? A 2007 study of 303 American students learning French found that grades were most strongly linked to confidence in managing one's own study, such as using time and learning methods well.

Confident learners tend to achieve more. But which comes first? Does confidence drive achievement, or does achievement build confidence? Educational psychology increasingly says both. In a 2011 review, Marsh and Martin supported a reciprocal effects model, in which earlier achievement raises academic self-concept and self-concept then predicts later achievement. These effects held for self-concept tied to a specific subject. General self-esteem had a negligible influence.

So how much do self-beliefs actually add? A 2004 meta-analysis of longitudinal studies by Valentine and colleagues controlled for earlier achievement and found the effect small but favorable. It was stronger when beliefs and outcomes were measured in the same domain.

In language learning, Mills, Pajares and Herron surveyed 303 American university students of French. The belief most closely tied to grades was self-efficacy for self-regulation, the sense that one can manage study time and use effective strategies. It mattered more than confidence in getting good results or low anxiety.

Their data came from a single point in time, however, so cause and effect cannot be separated. The confidence linked to grades in that survey was confidence in running one's own learning, and keeping a record of little successes is one way to build it.