The subject of this article is the extension of a study done by Rieben, Ntamakiliro, Gonthier and Fayol (2005) on the effectiveness of different writing practices among 5 year-old children. These researchers showed that in the psycholinguistic evaluation done at the end of the year, the results of the students who copied a series of words during the training phase were no different than those of the control group. This analysis studies the 6,048 words copied by these same students during their training sessions in an effort to contribute to an explanation of why copying these words had no effect. Three analyses were carried out: the number of words correctly copied, the frequency of the different types of errors, the relationships between the errors and performance in two psycholinguistic tests (knowledge of letters and phonemic detection). The results show that the copies present a good qualitative level and that there is no link between the frequency of the different types of errors and psycholinguistic test scores, except for the fact that the children who produce the most unrecognizable letters obtain a lower score in letter knowledge in the initial evaluation than children who produced the fewest unrecognizable letters. The results obtained do not allow a correlation to be drawn between the lack of effect the copying had to deficiencies in the copying itself. Therefore, other explicative hypotheses should be tested.