Florida Center for Reading Research, Florida State University, Tallahassee, FL, USA.
Behav Res Methods. 2024 Apr;56(4):2828-2841. doi: 10.3758/s13428-023-02205-2. Epub 2023 Aug 16.
Given English orthography's quasi-regular nature, applying common decoding rules to a word does not always result in a correct pronunciation matching the stored phonological form (e.g., the word tongue). To arrive at a correct pronunciation, developing readers must make the match between a decoded pronunciation and a word's correct pronunciation stored in memory. Developmentally, this matching process varies as a function of child skill (e.g., decoding, vocabulary) and word characteristics (e.g., spelling-to-pronunciation transparency, concreteness), with each being continuously distributed. Spelling-to-pronunciation transparency ratings represent a global measure of the ease of arriving at a word's correct pronunciation from its decoded pronunciation and in experimental studies has been shown to be a critical dimension in assessing the difficulty of a word for developing readers (e.g., Steacy et al., 2022a, 2022b). This study aimed to create a database of spelling-to-pronunciation transparency ratings for the 23,282 most frequently written English words, made available in the supplemental materials for future analyses. We asked adults to rate words' spelling-to-pronunciation transparency on a scale of 1-6 (1 = very easy to match, 6 = very difficult). Results of multiple regression analyses revealed variance in ratings to be unaccounted for by other word features, demonstrating the uniqueness of these ratings. Furthermore, words that are considered irregular, classified previously as strange, or contained at least one schwa received higher ratings, demonstrating strong associations between transparency and regularity. Lastly, these ratings significantly predicted both adult word naming time and child word reading accuracy above and beyond other word features known to predict reading.
鉴于英语拼写的准规律性，应用常见的解码规则并不总能得出与存储的语音形式匹配的正确发音（例如，单词 tongue）。为了得出正确的发音，发展中的读者必须在解码发音和存储在记忆中的单词正确发音之间建立匹配。在发展过程中，这个匹配过程因儿童技能（例如，解码、词汇）和单词特征（例如，拼写-发音透明度、具体性）而异，每个特征都是连续分布的。拼写-发音透明度评级代表了从解码发音到单词正确发音的难易程度的总体衡量标准，在实验研究中已被证明是评估发展中读者单词难度的关键维度（例如，Steacy 等人，2022a，2022b）。本研究旨在为最常书写的 23,282 个英语单词创建拼写-发音透明度评级数据库，这些评级可在补充材料中获取，供未来分析使用。我们要求成年人在 1-6 分的范围内（1=非常容易匹配，6=非常困难）对单词的拼写-发音透明度进行评级。多元回归分析的结果表明，这些评级的差异无法用其他单词特征来解释，这证明了这些评级的独特性。此外，被认为不规则的、被归类为奇怪的或至少包含一个中元音的单词的评级更高，这表明透明度和规律性之间存在很强的关联。最后，这些评级在考虑到已知可预测阅读的其他单词特征之外，显著预测了成人单词命名时间和儿童单词阅读准确性。
