Table 8. Hierarchical regression analyses summary for the variables predicting reading comprehension (n = 36)
Step and predictor variable
Reading accuracy (errors)
Cognitive Awareness Strategy Index
Metacognitive Awareness Strategy Index
* p < .05, ** p ≤ .001
In step 1, reading accuracy and reading speed were included as predictors in the regression analyses in order to control their effects on further steps. In step 2 of the first regression analysis, the addition of Cognitive Awareness Strategy Index to the equation resulted in a significant increment in multiple R2 (7.7%). In step 2 of the second regression analysis, the addition of Metacognitive Awareness Strategy Index to the equation resulted in a significant increment in multiple R2 (8.1%). Thus, both cognitive and metacognitive awareness contributed significantly to predicting reading comprehension, explaining additionally about 8% of the variance in reading comprehension, over and above that afforded by differences in reading accuracy and reading speed. Not unexpectedly, reading accuracy and reading speed explained together 35% of the variance in reading comprehension.
DISCUSSION AND CONCLUSIONS
The main goal of the present study was to establish an understanding of perceived use of cognitive and metacognitive reading strategies of poor and good readers. The interest was primarily focused on investigating possible differences in strategy awareness between poor and good readers, in terms of frequency and efficiency. Moreover, an attempt was made to examine whether the cognitive strategy awareness and metacognitive strategy awareness are important predictors of reading comprehension.
From the retrospective interview data, that shed light on readers’ strategy awareness, it can be concluded that the poor readers were able to describe the use of a number of cognitive strategies to the same extent with the good readers. However, according to verbal reports, poor readers employed cognitive strategies less frequently and less efficiently than good readers did. Good readers utilised more frequently meaning-oriented reading, while poor readers adopted a word-centred model of reading, tried to process word meaning rather than trying to comprehend and retain the meaning of the text. Thus, they reported less frequently certain ‘demanding’ cognitive strategies, such as guessing from the context, activating prior knowledge, using imagery, keeping meaning in mind, as well as strategies based on linguistic features of the text. Their reports, compared with those of good readers showed that they did not keep a proper balance between more and less sophisticated cognitive strategies. Our findings support previous studies which indicated that good readers are aware of their purposes for reading and employ repertoires of cognitive strategies for processing texts (Grabe & Stoller, 2002; Pressley, 2002), as well as they use context and prior knowledge more efficiently for comprehension purposes (Van Dijk & Kintsch, 1983; Kintsch, 1988).
Concerning metacognitive strategies, poor readers were aware of a smaller repertoire of metacognitive strategies, since they reported that they used metacognitive strategies less frequently than good readers did. Good readers were more aware that the reading tasks can require different approaches, they were selectively attentive as well as they were able to take a larger, more synthetic view; however, poor readers replied on a much slower analytical procedure. Clear and significant differences were found to exist between the two groups in relation to monitoring comprehension. Although, good readers were aware of employing text-processing strategies for monitoring comprehension, poor readers tended to employ word-level cues to focus on decoding the text and they did not frequently activate content schemata when needed, and did not control reading comprehension