The reciprocal relationships between literacy skills and various subscales of mental health have been examined in few studies. The objective is to re-examine this question in a longitudinal study with the possibility of a gender effect.
A 4-year longitudinal study was conducted as part of the French national cohort ELFE (N = 1064). The aim was twofold: 1/ To examine the links in both directions between literacy skills, assessed successively at 5, 7 and 9 years (LitSk5y; 7y; 9y), and mental health, assessed with the Strengths and Difficulties Questionnaire (SDQ) at 5 years by parents and 9 years by teachers. 2/To determine whether the gender effect on LitSk is still present when SDQ subscale scores show mental health difficulties. SDQ subscales consist of conduct problems, hyperactivity, emotional symptoms and peer relationships assessed at 5 years old, and conduct problems, hyperactivity and prosocial behaviour assessed at 9 years old.
Negative links were found in the following directions with SDQ subscales as predictors: conduct problems at 5 years old to LitSk5–7–9y and conduct problems at 9 years old to LitSk9y, and there was also a positive link, prosocial behaviour at 9 years old to LitSk9y. Conversely, negative links were again found with LitSk considered as a predictor of SDQ subscales: LitSk5y to conduct problems, hyperactivity and peer relationships at 5 years old, LitSk5–7–9y to conduct problems and hyperactivity at 9 years old; a positive link was found for LitSk5–7–9y to prosocial behaviour at 9 years old. The links for SDQ subscales to LitSk were stronger than those for LitSk to SDQ subscales.
These results revealed three bidirectional relationships: conduct problems at 5 years old with LitSk5y, conduct problems at 9 years old with LitSk9y and prosocial behaviour at 9 years old with LitSk9y. Finally, the expected LitSk gender effect in favour of girls was reduced when these latter were affected by mental health problems.