Gentaz Edouard, Sprenger-Charolles Liliane, Theurel Anne
Laboratoire de Psychologie et Neurocognition (CNRS) and Univ. Grenoble Alpes, Grenoble, France; Faculté de Psychologie et des Sciences de l'Education, Université de Genève, Geneva, Switzerland; LPNC, Centre national de la recherche scientifique, Grenoble, France.
Laboratoire de Psychologie Cognitive (CNRS), Aix-Marseille Université, Marseille, France.
PLoS One. 2015 Mar 20;10(3):e0119581. doi: 10.1371/journal.pone.0119581. eCollection 2015.
Based on the assumption that good decoding skills constitute a bootstrapping mechanism for reading comprehension, the present study investigated the relative contribution of the former skill to the latter compared to that of three other predictors of reading comprehension (listening comprehension, vocabulary and phonemic awareness) in 392 French-speaking first graders from low SES families. This large sample was split into three groups according to their level of decoding skills assessed by pseudoword reading. Using a cutoff of 1 SD above or below the mean of the entire population, there were 63 good decoders, 267 average decoders and 62 poor decoders. 58% of the variance in reading comprehension was explained by our four predictors, with decoding skills proving to be the best predictor (12.1%, 7.3% for listening comprehension, 4.6% for vocabulary and 3.3% for phonemic awareness). Interaction between group versus decoding skills, listening comprehension and phonemic awareness accounted for significant additional variance (3.6%, 1.1% and 1.0%, respectively). The effects on reading comprehension of decoding skills and phonemic awareness were higher in poor and average decoders than in good decoders whereas listening comprehension accounted for more variance in good and average decoders than in poor decoders. Furthermore, the percentage of children with impaired reading comprehension skills was higher in the group of poor decoders (55%) than in the two other groups (average decoders: 7%; good decoders: 0%) and only 6 children (1.5%) had impaired reading comprehension skills with unimpaired decoding skills, listening comprehension or vocabulary. These results challenge the outcomes of studies on "poor comprehenders" by showing that, at least in first grade, poor reading comprehension is strongly linked to the level of decoding skills.
基于良好的解码技能构成阅读理解的一种自引导机制这一假设,本研究调查了在392名来自低社会经济地位家庭的说法语的一年级学生中,与阅读理解的其他三个预测因素(听力理解、词汇和音素意识)相比,前一种技能对后一种技能的相对贡献。根据通过假词阅读评估的解码技能水平,这个大样本被分成了三组。使用高于或低于整个人口平均值1个标准差的临界值,有63名良好解码器、267名中等解码器和62名差解码器。阅读理解中58%的方差由我们的四个预测因素解释,解码技能被证明是最佳预测因素(12.1%,听力理解为7.3%,词汇为4.6%,音素意识为3.3%)。组与解码技能、听力理解和音素意识之间的相互作用分别解释了显著的额外方差(3.6%、1.1%和1.0%)。解码技能和音素意识对阅读理解的影响在差解码器和中等解码器中比在良好解码器中更高,而听力理解在良好解码器和中等解码器中解释的方差比在差解码器中更多。此外,阅读理解技能受损的儿童在差解码器组中的比例(55%)高于其他两组(中等解码器:7%;良好解码器:0%),只有6名儿童(1.5%)的阅读理解技能受损,但解码技能、听力理解或词汇未受损。这些结果对关于“阅读理解差的人”的研究结果提出了挑战,表明至少在一年级,阅读理解差与解码技能水平密切相关。