what the research says

Differentiated Instruction: Fine, but what to do with the rest of the class?

…use of reading groups creates a problem; when the teacher is working with one reading group, the other students in the class must be occupied with activities they can complete with minimal teacher direction. Research on these follow-up activities, or unsupervised seatwork, indicates that they are often of poor quality, are rarely taken seriously by teachers or students, and are poorly integrated with other reading activities (e.g., Beck, McKeown, McCaslin, & Burkes, 1979; Osborn, 1984), and that student time on-task during follow-up periods is typically low (e.g., Anderson, Brubaker, Alleman-Brooks, & Duffy, 1985). Yet in a class with three reading groups, as much as two thirds of the reading period is spent on follow-up activities.

Stevens, R.J., Madden, N.A, Slavin, R.E, & Farnish, A.M. (1987). Cooperative Integrated Reading and Composition: Two Field Experiments. Reading Research Quarterly. 22 (4) p. 433-454.

Interventions: small groups, 15-30 minutes-a-day is optimal

Although not conclusive for making individual student-level decisions, there is compelling research indicating that instruction provided to groups of 3 to 5 students is as effective as 1-to-1 instruction, even for the most at-risk students (Elbaum, Vaughn, Hughes, & Moody, 2000).

Harn,B.A., Linan-Thompson, S., Roberts, G. (2008). Intensifying Instruction: Does Additional Instructional Time Make a Difference for the Most At-Risk First Graders? Journal of Learning Disabilities, 41 (2) p.115-125.

And later in the same piece:

In a meta-analysis examining the role of time (both frequency and duration) on the effectiveness of kindergarten interventions, Cavanaugh et al. (2004) documented the great variability of intensity of time in effective interventions. In terms of intensity of time implementation, three of the five interventions that were implemented for fewer than 15 minutes per instructional session resulted in large effect sizes, 1.0 or higher. The majority of interventions were implemented from 15 to 29 minutes (n = 17) and had moderate to large effect sizes. The authors concluded that interventions delivered in small-group formats either two to three times per week or daily for a minimum of 15 to 30 minutes produced the greatest effects.

Harn,B.A., Linan-Thompson, S., Roberts, G. (2008). Intensifying Instruction: Does Additional Instructional Time Make a Difference for the Most At-Risk First Graders? Journal of Learning Disabilities, 41 (2) p.115-125.

When to use repeated reading as an intervention

Vadasy and Sanders ran a study that shows at-risk second and third graders benefit from repeated reading:

These findings suggest when particular types of repeated reading will beneficial. For students performing within the emergent stage of reading fluency, like the second and third graders in the study, there were benefits from rereading the same texts chosen at an instructional level of difficulty. Repeated reading was assisted by trained tutors who provided both a model of fluent reading during two of the rereadings, as well as corrections and feedback during three of the readings

Vadasy, P.F., & Sanders, E.A. (2008). Repeated Reading Intervention: Outcomes and Interactions With Reader Skills and Classroom Instruction. Journal of Educational Psychology, 100 (2), pp. 272-290.

Students at benchmark and above on word reading do not benefit:

… if students enter second or third grade with average or higher word reading skills, a repeated reading interventions is not recommended to boost word reading accuracy or comprehension skills in particular. Students with lower word reading skills at onset, however, will likely benefit from this type of intervention on reading accuracy and fluency rate gains

Vadasy, P.F., & Sanders, E.A. (2008). Repeated Reading Intervention: Outcomes and Interactions With Reader Skills and Classroom Instruction. Journal of Educational Psychology, 100 (2), pp. 272-290.

How kids learn to summarize text

Researchers Brown and Day examined the developmental sequence of six strategies students use to summarize text (my emphasis):

Two of the six rules involve deletion of unnecessary material. One should obviously delete material that is trivial, and even grade school children are quite adept at this if the form and content of the material is familiar (Brown, Day, & Jones, in press; Johnson, 1978). One should also delete material that, although it is important, is also redundant. Kintsch and van Dijk’s system also includes these two deletion rules. Two of the rules of summarization involve the substitution of a superordinate term or event for a list of items or actions. For example, if a text contains a list such as: cats, dogs, goldfish, gerbils, and parrots, one can substitute the term pets. This is Kintsch and van Dijk’s generalization rule. Similarly, one can substitute a superordinate action for a list of subcomponents of that action, i.e., John went to London, for: John left the house, John went to the train station, John bought a ticket, etc., etc. This is roughly comparable to Kintsch and van Dijk’s (1978) integration rule. The two remaining rules have to do with providing a summary of main constituent unit of text, the paragraph. The first rule is – select a topic sentence, if any, for this is the author’s summary of the paragraph. The final rule is – if there is no topic sentence, invent your own.

Brown, A.L., & Day, J.D. (1983). Macrorules for Summarizing Texts: The Development of Expertise. Center for the Study of Reading, Illinois University

What they found may help teachers identify the developmental stage a student is at:

“…a clear developmental pattern was found, with deletion rules emerging first followed by superordination [substitution] and then selection. Invention, the most difficult rule, was late developing. We believe that the five rules differ in their ease of application because they demand different degrees of text manipulation on the part of the learner, and perhaps because they depart to a greater or lesser extent from the already existing strategy favored by the younger participants. This has been called the copy-delete strategy (Brown, 1981; Brown, Day, & Jones, in press) because fifth and seventh grade and junior college students summarize texts primarily by deleting, or copying near verbatim the words actually present in the text

Brown, A.L., & Day, J.D. (1983). Macrorules for Summarizing Texts: The Development of Expertise. Center for the Study of Reading, Illinois University

Intriguingly, they suggest that an earlier, over-simple strategy may need to be actively unlearned in order to progress to the later strategies:

““…there is evidence that partially adequate strategies such as copy-delete are not just way-stations on the road to expert strategies; they may actually impede progress...We would like to argue that partially adequate strategies such as copy-delete and knowledge-telling are maintained by inexperienced writers because they do result in intermittent reinforcement and are recognizable attempts to get the job done. The process of development is not just one of acquiring increasingly more refined and sophisticated strategies; development involves the systematic consolidation of mature strategies, combined with the rejection of plausible but less efficient habits

Brown, A.L., & Day, J.D. (1983). Macrorules for Summarizing Texts: The Development of Expertise. Center for the Study of Reading, Illinois University

Sound + vision (phonology and orthography)

A great deal of research shows that average reading ability and below-average reading ability are best predicted by the somewhat independent effects of phonologically based decoding and rapid naming (Compton, DeFries, & Olsen, 2001; Cornwall, 192; Kirby, Parrila, & Pfeiffer, 2003; Savage et al., 2005; Strattman & Hodson, 205; Wolf & Bowers, 1999; Wolf & O’Brien, 2001).

Savage, R., Pillay, V., & Melidona, S. (2008). Rapid Serial Naming is a Unique Predictor of Spelling in Children. Identifying and Predicting Classes of Response to Explicit Phonological Spelling Instruction During Independent Composing. Journal of Learning Disabilities 41 (3), pp. 235 -250

The neuroscience suggests that "storing the picture" of a word and "processing the sounds" in it are two somewhat-independent pathways. A student could have a deficit in either or both, making diagnosis more complicated and adding weight to the idea that you have to be "firing on all cylinders" to read and comprehend successfully.

In current models, RAN [Rapid Automatized Naming] is seen as a measure of an underlying structure or timing congruence in the central nervous system that underlies the capacity to accurately and efficiently store detailed orthographic information about words. This structure is seen as being somewhat independent of the structures used for phonological processing (e.g., Wolf, Bowers, & Biddle, 2000).

Savage, R., Pillay, V., & Melidona, S. (2008). Rapid Serial Naming is a Unique Predictor of Spelling in Children. Identifying and Predicting Classes of Response to Explicit Phonological Spelling Instruction During Independent Composing. Journal of Learning Disabilities 41 (3), pp. 235 -250

Student books are too difficult

A sea change occurred in text characteristics used in reading textbooks between the late 1980s and early 1990s. Textbook programs for beginning and developing readers that had been controlled by readability formulas were replaced by children’s literature (Hoffman et al., 1994). With the use of literature in textbook programs rather than controlled texts, the number of total words in texts decreased while the number of unique words increased (Hoffman et al., 1994). Authors of children’s literature tend to limit the plots and story length of books for children, but children’s literature contains substantially more rare words than either earlier school texts or typical adult conversations (Hayes, Wolfer, & Wolfe, 1996). By the late 1990s, several states were calling for textbook reform. Texas (Texas Education Agency, 1997) and California (California English/Language Arts Committee, 1999) mandated that reading textbook programs purchased with state monies include decodable texts.

Hiebert, E.H., and Fisher, C.W. (2002, May). Text matters in developing fluent reading. Paper presented at the annual meeting of the International Reading Association, San Francisco, CA

So far, so good...but:

Several years after these mandates, literature continued to dominate textbook programs beyond the first semester of first grade (Hiebert, 2001). Further, first-grade texts continued to feature high numbers of unique words, many of which appear a single time (Foorman, Francis, Davidson, Harm, & Griffin, 2002).

Hiebert, E.H., and Fisher, C.W. (2002, May). Text matters in developing fluent reading. Paper presented at the annual meeting of the International Reading Association, San Francisco, CA

Can school systems provide the intervention resources students need?

Although early intervention using evidence-based practices may be widely promoted, implementation in real classrooms and schools is challenging (Abbott, Greenwood, Buzhardt, & Tapia, 2006; Fixen, Naoom, Blase, Friedman, & Wallace, 2005). Challenges center on pedagogy, resources, and changing systems to support early reading intervention, particularly in kindergarten through second grade.”

Kamps, D., Abbott,D., Greenwood, C., Wills, H. et al. (2008). Effects of Small-Group Reading Instruction and Curriculum Differences for Students Most at Risk in Kindergarten: Two- Year Results for Secondary- and Tertiary- Level Interventions. Journal of Learning Disabilities,  41 (2) p.101- 114.

Small group interventions 15 to 30 minutes long get greatest effects

In a meta-analysis examining the role of time (both frequency and duration) on the effectiveness of kindergarten interventions, Cavanaugh et al. (2004) documented the great variability of intensity of time in effective interventions. In terms of intensity of time implementation, three of the five interventions that were implemented for fewer than 15 minutes per instructional session resulted in large effect sizes, 1.0 or higher. The majority of interventions were implemented from 15 to 29 minutes (n = 17) and had moderate to large effect sizes. The authors concluded that interventions delivered in small-group formats either two to three times per week or daily for a minimum of 15 to 30 minutes produced the greatest effects.

Harn, B.A., Linan-Thompson, S., Roberts, G. (2008). Intensifying Instruction: Does Additional Instructional Time Make a Difference for the Most At-Risk First Graders? Journal of Learning Disabilities, 41 (2) p.115-125.

Small groups of 3-5 are as effective as 1-to-1 instruction

Although not conclusive for making individual student-level decisions, there is compelling research indicating that instruction provided to groups of 3 to 5 students is as effective as 1-to-1 instruction, even for the most at-risk students (Elbaum, Vaughn, Hughes, & Moody, 2000).

Harn, B.A., Linan-Thompson, S., Roberts, G. (2008). Intensifying Instruction: Does Additional Instructional Time Make a Difference for the Most At-Risk First Graders? Journal of Learning Disabilities, 41 (2) p.115-125.

Scripted vs explicit instruction

To illustrate the difference in scripted versus explicit, Torgesen et al. (2001) compared the efficacy of a highly scripted reading program with a clearly defined scope and sequence with an approach that taught the alphabetic principle explicitly but spent more time reading and writing in context. There were no significant differences in outcomes for a group of elementary school children with severe reading disabilities (see also Wise, Ring, & Olson, 2000).

Stuebing, K.K., Barth, A.E, Cirino, P.T., Franis, D.J., et al. (2008). A Response to Recent Reanalyses of the National Reading Panel Report: Effects of Systematic Phonics Instruction Are Practically Significant. Journal of Educational Psychology, 100 (1), p. 123-134

IOW, tight scripting does not provide an advantage over explicit-but-not-scripted instruction. It's the explicitness that provides the benefit, not the scripting. More evidence of the same:

Mathes et al. (2005) compared two comprehensive small-group tutorial interventions based on (a) a direct instruction model with a scripted lesson plan and a well-developed scope and sequence with use of decodable text; and (b) a guided reading intervention in which instruction in the alphabetic principle was explicit (i.e., based on a plan for introducing phonics elements and in which the information was directly presented to the child) and done for about 20% of the instructional period, but unscripted and with the use of leveled texts instead of decodable texts. No major differences in reading outcomes for first graders at risk for reading difficulties were apparent when these two comprehensive programs were compared.

Stuebing, K.K., Barth, A.E, Cirino, P.T., Franis, D.J., et al. (2008). A Response to Recent Reanalyses of the National Reading Panel Report: Effects of Systematic Phonics Instruction Are Practically Significant. Journal of Educational Psychology, 100 (1), p. 123-134