
Most spelling errors made by elementary school students are associated with the production of multigraphemic phonemes and silent letters. To document the best ways of teaching these spelling phenomena, two experimental conditions were tested with two groups of 7-8 year-old students: 1) teaching formal and semantic word properties (n = 43) and 2) teaching formal word properties (n = 44). A third group of students constituted the control group (n = 44). For both experimental conditions, the same 24 words were taught to the students. To assess the students’ learning, they were given a dictation before and after the intervention. The results show that the intervention contributed to the learning of multigraphemic phonemes and silent letters for students who benefited from the experimental conditions tested, although variations were observed depending on the orthographic phenomena.