Early Reading Instruction: What Science Really Tells Us about How to Teach Reading

McGuinness shows that all writing systems, without exception, are based on a sound unit in the language. This fact, and other findings by paleographers, provides a platform for the prototype. Other elements of the prototype are based on modern research. For example, observational studies in the classroom show that time spent on three activities strongly predicts reading success: learning phoneme/symbol correspondences, practice at blending and segmenting phonemes in words, and copying/writing words, phrases, and sentences. Most so-called literacy activities have no effect, and some, like sight word memorization, have a strongly negative effect.

The National Reading Panel (2000) summarized the research on reading methods after screening out thousands of studies that failed to meet minimum scientific standards. In an in-depth analysis of this evidence, McGuinness shows that the most successful methods (children reading a year or more above age norms) include all the elements in the prototype. Finally, she argues, because phonics-type methods are consistently shown to be superior to whole-word methods in studies dating back to the 1960s, it makes no sense to continue this line of research. The most urgent question for future research is how to get the most effective phonics programs into the classroom.