More on Teaching Reading

What's amazing to me about the whole enterprise of teaching reading is that there are still some stalwarts of the "whole language" method out there, holding the faith in their discredited belief that de-emphasizing or eliminating phonics instruction is the best way to teach reading.

A recent study: Patricia F. Vadasy, Elizabeth A. Sanders, and Julia A. Peyton, "Code-Oriented Instruction for Kindergarten Students At Risk for Reading Difficulties: A Randomized Field Trial With Paraeducator Implementers," Journal of Educational Psychology Vol. 98, No. 3 (2006): 508-28.

Randomized field trials, to my knowledge, are relatively rare in the educational literature. But they are useful, because in other types of studies (i.e., where two different schools just happen to use different curricula), it can be more easily alleged that any difference in the results is really due to unobserved differences between the two schools, selection effects, etc.

What the researchers found was that the groups that received "18 weeks of explicit instruction in phonemic skills and the alphabetic code" then "significantly outperformed controls on measures of reading accuracy, reading efficiency, oral reading fluency, and developmental spelling." Significant group differences remained at a one-year follow-up, so much so that the researchers pointed out that "refraining from providing intervention to students in the follow-up stage of this study would have had negative social consequences." In other words, it was akin to a drug trial where the researchers find that a heart attack drug is so helpful that it's no longer ethical to let the control group go untreated.

The researchers also included this helpful summary of the literature:

Interventions have been studied extensively to show that classroom teachers, when provided literacy curricula that include explicit and well-designed phonemic awareness and alphabetic acativities along with training and professional development, enable at-risk kindergarten students to attain a positive trajectory of development in critical early reading skills (Coyne, Kame'enui, Simmons, & Harn, 2004; Elbro & Petersen, 2004; Foorman et al., 2003; Fuchs et al., 2001; Torgesen, Wagner, Rashotte, et al., 1999). Effect sizes for these teacher-implemented interventions have been robust. Typical effect sizes in decoding have been 0.55 (Elbro & Petersen, 2004), 0.79 (Fuchs et al., 2001), 0.92 (Simmons, Kame'enui, Stoolmiller, Coyne, & Harn, 2003), and 0.57 (Torgesen, Wagner, Rashotte, et al., 1999). . . . For example, a meta-analysis of 36 training studies on phonological and letter-sound skills (Bus & van Ijzendoorn, 1999) reported average phonological and reading effect sizes in kindergarten of d = 1.26 and d = 0.32, respectively.