The point of mastering the procedural knowledge of decoding text, recoding text, practicing decoding to facilitate the word superiority effect (Reicher, 1969) to ensure decoding automaticity and the further practice and knowledge and vocabulary harvesting that permits reading fluency is to be able to understand the meaning that an author wishes to communicate. Reading fluency gifts the working memory the release of capacity to concentrate its limited resources on reading comprehension.
The impression given is that reading comprehension is merely the next stage in a hierarchical structure which can be taught, much as reading fluency can be achieved through instruction and practice.
The embedding of the concept that reading comprehension is a skill dependent on the learning of a series of strategies that can, and should, be taught, has its roots in the US National Reading Panel report (NICHD, 2000) that identified seven comprehension strategies that had sufficiently strong research bases to warrant their adoption with the implied suggestion that the teaching of comprehension involved the teaching of these strategies.
The research base was first established in the 1970s and 80s with single strategy instruction studies. Students were taught individual comprehension strategies (summarising, creating mental images of the text, activating prior knowledge) with the results favouring those students taught the strategies (Tierney and Cunningham, 1984) and interpreted as evidence that students could be taught reading comprehension. This was followed in the 1980s by research into the effects of teaching students the application of multiple strategies where a repertoire of strategies was taught. Again, the positive effects on researcher-devised tests (although less emphatic on standardised tests) was interpreted as evidence that students could be taught multiple strategies simultaneously (Duffy et al., 1987). This was developed further with the introduction of Collaborative Strategic Reading (Kim et al., 2006) where small heterogenous peer groups were assigned different roles within the collaboration that aligned to the relevant strategies. The apparent success of this approach led to the development of Transactional Strategies Instruction (Pressley et al., 1992) that focused on a repertoire of strategies that concentrated on transactions among participants for the joint construction of meaning.
There seemed to be no doubt that the teaching of comprehension strategies directly improved comprehension.
There was, however, a gnawing doubt.
The ability to read every word ensures that the mind is assailed by every word. Thus, the mind is charged with determining which words to repress, emphasise, correlate and organise as it constructs the right weight, influence and force for each expression, all under the influence of establishing purpose and meaning from a non-present communicator (Thorndike, 1917). This attempt to simultaneously extract and construct meaning (RAND, 2002) requires an interplay between the reader’s knowledge, capabilities and sociocultural context (Wilkinson and Son, 2013). This results in meaning, the product of comprehension, never being stable. A change in one variable, knowledge for example, will alter the construction of meaning with the product of comprehension changing moment by moment (Pearson, 2001).
This integration of the text base with the reader’s knowledge base to construct understanding implies that reading comprehension is a dynamic, constantly fluctuating phenomena that is a context sensitive process (Kintsch, 1998). Prior knowledge informs the construction of the text base and the new knowledge then forms part of the reader’s long-term memory store for use in understanding new texts. Every time a new text is read, it is informed by previous reading or, ‘Knowledge begets comprehension begets knowledge’ (Pearson, 2006).
This seems at odds with the research (NICHD, 2000) that indicated that comprehension was a skill that could be taught by utilising strategies and that reading comprehension was a static, stable phenomena (Pressley, 2006).
The explicit teaching of comprehension strategies had raised concerns before the NRP report with the questioning of whether teaching children how they think they could improve their thinking (Pearson and Gallagher, 1983) and whether a focus on process at the expense of understanding would undermine that understanding (Cunningham, 1984). Pearson and Fielding (1991) suggested that the explanations and self-reflections required by explicit strategy instruction might be more complicated than the task of understanding what the author actually meant. They concluded with the then radical suggestion that the need for comprehension strategies could be eliminated by focusing on the content of the text itself
Whether a skilled reader has time to utilise individual reading strategies was raised by Resnick (1985) who suggested that the sheer pace and automaticity of an accomplished reader precluded the adoption and application of individual strategies whilst reading. This was supported by Rosenshine and Meister (1994) whose review of research indicated that comprehension was generally the same regardless of which strategies were taught and that there was no relationship between the ability to use the strategy of generating questions and students’ ability to comprehend. Furthermore, there was a suggestion that the use of and reliance on reading strategies diverted cognitive resources from the more fundamental task of deriving meaning from the text and that there was no one strategy that appeared the most effective (Sinatra et al., 2002).
So why does the teaching of comprehension strategies appear to work better than a deep dialogic analysis of the text?
It doesn’t.
When strategies instruction was compared with instruction that focused on text content in response to meaning-based questions (McKeown, Beck and Blake, 2009; Garcia et al., 2007) there was no significant difference in performance of the two groups. Where there was a significant difference was between the control groups who received no instruction, whose performance was substantially inferior. It would seem that it is the active engagement with the text as a result of strategy instruction that is at the heart of its success (Kintsch and Kintsch, 2005).
So, some instruction is better than none. You pays your money and you takes your choice. And as strategy teaching is so much easier than text analysis that requires expert knowledge of a text and the components of it so we should use strategies?
Not necessarily.
The problem is the teaching and the teacher. Strategy instruction often results in the strategy becoming the focus and an end point rather than a means to an end (Paris and Winograd, 1990; Tierney and Cunningham, 1984) with teachers overemphasising the strategy to the detriment of students engaging in a meaningful dialogue with the text (Marks et al., 1993) and ignoring the vegetables and becoming, ‘more interested in the tools in the shed…’ (Tierney and Cunningham, 1984 p. 634).
A further problem is that although research studies indicated significant improvements in comprehension as a result of strategy instruction this is not evident in the wider world of day to day teaching. Teachers find it difficult to sustain the instruction in class rooms with very little instruction observed, reducing to one minute a day (Connor et al., 2004) in many settings and what masquerades for comprehension instruction being little more than regular comprehension tests (Durkin, 1979).
Strategy instruction has also had its greatest effects when the studies have involved students with learning disabilities (Gersten et al., 2001) and there is some evidence that although strategy instruction delivers fast improvement and good results in statutory testing at primary phases, these improvements are not sustained and result in catastrophic deficiencies in later schooling (Kozol, 2006).
It is pertinent that content-rich instruction for reading comprehension has proven particularly successful in upper primary phases in the comprehension of science texts (Guthrie et al., 2004) with effect sizes up to 0.93 on standardised comprehension tests. This compelling evidence suggests that reading comprehension is indeed a function of knowledge and vocabulary and that Willingham (2007) and Hirsh (2016) may be right: explicit instruction of relevant knowledge and the gleaning of a wide knowledge base is crucial to understanding text.
If comprehending text is indeed a dynamic and content sensitive cognitive activity, then perhaps we should look to address this in our teaching. The content teaching may be the easier to attain.
More on addressing the dynamic nature of deriving meaning next time...