Swanson and colleagues observed reading instruction in 50 elementary classrooms with 270 Spanish-speaking English-language learners (ELLs) over a period of two school years. Three grade levels were included in the study (first, second, and third). The researchers tracked the amount of time spent on six different teacher activities during reading instruction:
- grammar and spelling (rules of grammar, capitalization, punctuation, mechanics, and spelling)
- writing instruction
- phonemic awareness
- reading/listening (including “previewing books with children, book/print awareness, and spelling in context”)
- oral language, and
- vocabulary & structural analysis (related to “definitions and meanings of words and structures of words”) (p. 201-202).
Students were given a series of tests in Year 1 and Year 2. The tests included word attack (reading nonsense words), letter-word identification, and “passage comprehension” tests from the Woodcock Reading Mastery Test and the Woodcock-Muñoz Language Survey. The passage comprehension measure was a cloze assessment that “required children to identify specific words that went in the blank spaces of various passages” (p. 202). It was given in English.
The researchers found that the “best fit” for their statistical model predicting scores on the passage comprehension measure in Year 2 was one that included the amount of time teachers spent on grammar and spelling in Year 1, regardless of the grade level observed. The amount of time spent in Year 1 on phonemic awareness (PA) instruction did not predict performance on any of the Year 2 measures, even though these are the kinds of measures that PA training might influence (see Krashen & Hastings, 2011 for more details).
As has been true in many similar studies of this sort, no actual reading comprehension measures were administered. As we’ve previously discussed, the passage comprehension cloze assessments of the sort found in Woodcock-Muñoz are in fact mostly measures of word recognition, and are heavily influenced by skills instruction (Hua & Keenan, 2017, paywall).
The study provides no evidence that spelling and grammar instruction has any significant impact on the only measure that really matters in reading instruction – how well students are able to understand what they’re reading.
The answer to the question posed by the title of this article is therefore “We couldn’t really tell you because we didn’t actually measure it.”
Follow Us!