The aim of this study is to investigate the effect of more or less decodable texts (i.e., texts containing only graphemes already taught) on the performance of first graders. The use of decodable texts at the beginning of reading instruction is sometimes recommended (e.g., Castles et al., 2018; HMSO,2005). However, texts not fully decodable present a wider choice of words, which helps to maintain motivation in reading, and to develop students' comprehension (Price-Mohr & Price, 2019), vocabulary and knowledge (Solity & Vousden, 2009). Research is needed to better determine best practices. Reading, phonological awareness, spelling, comprehension and motivation of 246 French children from 15 first grade classrooms were assessed at the beginning of first grade (September) and re-assessed two months later. During this 2-months period, considering the graphemes and the words already taught, the mean percentage of decodable material read was calculated for each classroom. A multi-level analysis was performed to see whether and how this percentage influences the early acquisition of literacy skills, independently of classroom effect and early abilities effects. Descriptive analyses showed a high variability of mean percentage of decodable material read across classrooms, despite all the teachers used a phonic method for teaching to read. Three patterns of decodability emerged: classes always using moderately decodable texts, classes always using highly decodable texts, and classes offering increasingly decodable texts. The analyses revealed that the decodability patterns did not significantly affect students' performance in reading, phonological awareness and spelling after 10 weeks of instruction. The only significant effect was in line with Price-Mohr & Price (2019), that less decodable texts would promote comprehension. Further studies are needed to confirm these results.