(at another presentation)
Live blogging here in McKim Hall at the University of Virginia as Steve Graham delivers the McGuffey Reading Center’s 25th annual Graham Lecture. After Marcia Invernizzi’s cordial introduction, Steve began with a joke and a couple of humorous anecdotes about students’ writing. Of course, he tipped his hat to his collaborators and the sponsor of the research (Carnegie’s Writing Next).
Steve went into a rationale for the importance of writing instruction (“Why do reading, math, science, and technology get all the attention?)”. He then discussed forms of research, explaining that he was going to draw on experimental and quasi-experimental research, single-subject studies, and examinations of successful teachers. In addition, he noted that when the results from studies from diverse methods align, he has increased confidence in the strength of this findings.
After quick coverage of meta-analysis and of standards for judging effect sizes, he moved into a series of points about employing evidence-based practices. For the first one, he introduced an example of a mnemonic for teaching planning of written products (“STOP: Suspect judgment; Take a side; Organize ideas; Plan more as you write.”). [Steve, who doesn’t usually wear a jacket, has lost the one he was wearing during the introduction.] He used this to talk about the importance of getting students to think strategically about writing and then introduced results from a series of studies examining how writers think about writing and revising—students with disabilities focus revisions on mechanics and it takes a lot to get them to revise for concepts.
Steve moved from planning to talking about summarization. He listed a set of rules for summarization and discussed how to make them simple and more orderly.
For a third evidence-based practice (ES = .75 from 7 studies), Steve described peer-assistance strategies. He started with a funny story from his own experience. For the fourth practice, setting practice goals, five studies yielded an average ES of .70. That’s not too shabby, even though it’s based on few studies.
How about word processing? That’s Steve’s fifth evidence-based practice. He reported that 18 studies yielded an average ES of .55. He noted that the effects are important, but they need to be considered in the context of the need for strategic, planned writing.
Steve then moved into a sixth practice, sentence combining (5 studies; average ES=.50). He illustrated how sentences can be combined in myriad ways, not just one specific way. He illustrated additionally how multiple sentences can be combined differently,
[Low battery; I may have bail.]
Then there was the seventh practice: Process approach (21 studies; average ES .32). Steve pointed out that there were enough studies to permit sub-analyses. One of those revealed that teachers with substantial training in process approaches have greater effects (was the average .52?) than those who don’t have such training. Also, when additional studies where the measurements’ trustworthiness are not established are included, the ES goes lower.
8. Pre-writing activities: 5 studies with an average ES of .32.
9. Inquiry- [bailed].
There’s more! I just need to recreate it. Meanwhile, you can learn a lot of this by simply reviewing the Writing Next PDF from the commendable Alliance for Excellent Education site. Of course, one can jump back to my recent entry on the Graham-Perin meta-analysis right here on TE.