Independent Reading Levels are Problematic, Too

Shanahan response:

There is a body of research that explores the impact of reading on various outcomes and there is a body of research that explores the impact of teaching students to read with texts of certain levels, but I know of NO research into the impact of students' reading experiences at an independent level. 

There are several studies, however, that show students prefer NOT to read materials at their independent level (I know of only one exception to this). That is, even good readers tend to be interested in the subject matter and treatments of information that are harder than reading experts claim they should be reading. Thus, it is likely that the studies that have considered the impact of reading on various outcomes are not measuring the impact of the reading of independent level texts.

The most direct test of the effect of reading on learning was provided by a study by Carver and Liebert. They found no clear benefit resulting from 60 hours of additional reading for students even though the texts were presumed to be at the students’ levels. Given the failure of the approach, they hypothesized that more challenging texts may have been more effective. Unfortunately, no one has followed up on that.

There is a small body of research suggesting that having students read more at home or during the summer can improve reading achievement (a very small amount--less than 1 month on an elementary grade standardized test). I would suggest the research of Richard Allington & Ann McGill-Frantzen, or James Kim. But these studies do not measure the student-text match, so it would be impossible to conclude that students should practice at their independent level from that evidence.  

It seems clear that reading more is a good idea for most kids, and yet, we have no empirical data on which to base claims about how much reading or how challenging that reading should be.

Finally, the best research journals in reading (e.g., Scientific Studies of Reading, Reading Research Quarterly, Journal of Educational Psychology) are not easy for teachers to read. Journals like the Reading Teacher are more readable, but they usually do not publish research.  Teachers either need to learn actual research, or they must depend on second-hand sources that may or may not represent the original research accurately.