High and Low Reading Comprehension Achievers' Strategic Behaviors and Their Relation to Performance in a Reading Comprehension Situation

Abstract

This study aimed at investigating the actual strategic behaviors of high and low achievers in reading comprehension and their relation with respective performance. The participants were 45 individually examined third graders, 20 high and 25 low reading comprehension achievers. Cognitive, metacognitive, and motivational aspects of the participants' strategic behaviors during text comprehension was directly recorded by two observers by means of a structured observation form. The results revealed a relatively different profile of strategic behaviors between high and low achievers. Students' strategic behaviors during dealing with text comprehension were strongly related to respective performance for both groups.

Previous research has documented the importance of self-regulatory processes to improved reading and reading comprehension. Engaged, self-regulated readers are those who set realistic goals, select effective reading strategies, monitor their understanding of the text, evaluate progress toward their goals, and are highly motivated to read and learn; in other words, they actively use their own skills to build understanding (Gourgey, 2001; Pressley, 2002; Zimmerman, Bonner, & Kovach, 1996). Within this approach, the importance of students' strategic behaviors that facilitate active, independent, self-regulated reading is stressed more than before (Alexander, Graham, & Harris, 1998; Weinstein, Husman, & Dierking, 2000). Although reading comprehension depends on various text, situational, and student variables, mastering and applying appropriately multiple strategies and appropriate learning behaviors in reading discriminate skilled from less skilled readers (Cross & Paris, 1988; Jacobs & Paris, 1987; Magliano, Trabasso, & Graesser, 1999; Pressley, 2002; Pressley & Hilden, 2006).

It has been proposed that not only the knowledge and use of learning strategies are essential for learning and achievement but that various individual characteristics of learners influence their ability to be self-regulated and to act strategically during learning. Prior achievement, domain-specific knowledge, performance and regulation of tactics and strategies, and other cognitive and motivational individual differences are factors that affect learners' strategic behaviors and their monitoring and control processes (Alexander et al., 1998; Pressley & Hilden, 2006; Wang & Peverly, 1986; Winne, 1996). The present study was intended to investigate the strategic behavior of high and low reading comprehension achievers to perform and regulate text comprehension as it emerges within a specific text comprehension situation. Moreover, it was intended to examine the pattern of relations between the strategic behaviors examined and subsequent performance in reading comprehension in the groups of high and low reading comprehension achievers. There are not many direct studies of strategic learning as an individual difference variable (Winne, 1996). Moreover, recently, Siegler (2006) underlined the contribution of microgenetic analyses in the study of learning. This study examines the students' strategic behaviors “in action”; i.e., as they unfold during engagement in reading comprehension tasks, considering them as external indicators of self-regulatory processes. The vast majority of previous research examining students' self-regulation used mainly self-reports instrumentation in a domain of knowledge, such as self-completed questionnaires or verbal reports on the problem-solving process (Stokking & Voeten, 2000; Winne & Perry, 2000). Few studies provided data on students' actual self-regulatory behavior as it spontaneously emerges during a specific learning and problem-solving situation (see, for example, Juliebö, Malicky, & Norman, 1998; Whitebread, Coltman, Anderson, Mehta, & Pasternak, 2005).

Moreover, since self-regulation of learning is a complex construct and it is consisted of various components and processes (Paris & Paris, 2001), few studies investigated at the same time various categories of self-regulatory strategies that students use during problem-solving (see, for example, Harris & Graham, 1992). In the present study, different cognitive, metacognitive, and motivational aspects of students' reading and reading comprehension strategic behaviors were recorded in order to delineate the strategic profiles of high and low achievers in reading comprehension. In the next pages, literature on students' use of strategies is reviewed and the attributes of good and poor readers' strategic behaviors are outlined.

Self-Regulatory Strategic Behavior and Academic Performance

An essential aspect of self-regulated learning is the knowledge and use of learning strategies. Broadly defined, “… students' learning strategies include any thoughts, behaviors, beliefs, or emotions that facilitate the acquisition, understanding, or later transfer of knew knowledge and skills” (Weinstein et al., 2000, p. 733). Strategies are skills and they consist of various tactics. Winne (1996) stated that “… a strategy is a structured set of alternative tactics for navigating a task” (p. 328). Although strategies are considered as more conscious, intentional, and effortful than skills, “the same procedures that sometimes warrant the label strategies can also fit within the category of skills” (Alexander et al., 1998, p. 136). Therefore, it is not always easy to make clear conceptual distinctions between them.

During the last 20 years, research has emphasized the interaction of cognitive, metacognitive, and affective factors in learning (Efklides, Niemivirta, & Yamauchi, 2002; Gourgey, 2001; Paris & Oka, 1986). In literature, cognitive, metacognitive, and motivational processes and strategies are frequently identified (Mayer, 1998; Sperling, Howard, Miller, & Murphy, 2002; Weinstein et al., 2000; Winne, 1996; Wolters & Pintrich, 1998; Zimmerman, 1999). Cognitive strategies are the various rehearsal, elaboration, and organizational strategies, such as analyzing and combining activities and choosing between main and trivial information (Mayer, 1998; Pintrich, 1999; Pressley & Hilden, 2006; Weinstein et al., 2000; Winne, 1996; Wolters & Pintrich, 1998). Cognitive strategies enable students to make progress and to build knowledge. The use of cognitive learning strategies is connected to other aspects of self-regulation such as metacognition and motivation (Weinstein et al., 2000). Metacognitive and regulatory processes enable one to monitor, control, and evaluate performance; in other words, to use knowledge strategically (Corno, 1986; Gourgey, 2001; Mayer, 1998; Pintrich, 1999; Wolters & Pintrich, 1998; Zimmerman, 1999). Goal-setting, planning, self-monitoring, regulation of the cognitive process, and evaluation of the learning outcome are broader types of strategies that may serve as general or task-specific procedures (Alexander et al., 1998; Mayer, 1998; Paris & Paris, 2001; Winne, 1996; Wolters & Pintrich, 1998). Research has shown that strategic learners flexibly shift between cognitive and metacognitive activities while engaged in learning tasks (Butler, 1998).

Furthermore, strategies for monitoring and controlling one's own motivation toward the task to be solved, such as self-reinforcement, are also viewed as a critical determinant of students' self-regulated learning and achievement (Corno, 1986; Mayer, 1998; Meece, 1994; Sansone & Thoman, 2005; Weinstein et al., 2000; Zimmerman, 1999). Some motivational attributes of effective learners identified in recent literature are interest, curiosity, initiatives and high levels of students' activation, autonomous work, persistence in the face of difficulties, and maintaining self-motivation toward the task at hand (Kuyper, van der Werf, & Lubbers, 2000; Sansone & Thoman, 2005; Wolters & Rosenthal, 2000).

Several studies have consistently shown that students' application of various categories of strategies facilitates engaged, self-regulated learning and this may be directly related to their academic performance. There is powerful evidence from previous studies of the causal relationship between comprehension strategy use and comprehension (see Gourgey, 2001; Pressley, 2002). As regards students' cognitive and metacognitive skillfulness, it has been claimed that this skillfulness makes a significant contribution to the development of children as learners and to their academic achievement (Alexander et al., 1998; Brown, Bransford, Ferrara, & Campione, 1983; Gourgey, 2001; Pintrich & DeGroot, 1990; Pressley, 2002; Pressley & Hilden, 2006). Various instructional programs focused on instructing students to use a repertoire of comprehension strategies in a self-regulated fashion and they made a difference in the understanding of text (see Pressley & Hilden, 2006). As regards evidence on the relations between students' activities for retaining self-motivation and for regulating motivation and academic performance, the results are not conclusive, since motivational regulation variables in some studies were directly related to academic skills and performance (e.g., Onatsu-Arvilommi, Nurmi, & Aunola, 2002), whereas in other studies they were indirectly related to performance (e.g., Kuyper et al., 2000). Past literature showed that motivational factors mainly affect students' performance indirectly through their influences on intention to learn, persistence, quality of cognitive engagement, use of strategies, and estimation of learning situation.

Self-Regulatory Strategies in Reading Comprehension: Attributes of Good and Poor Readers

The three broad categories of strategies presented in the previous section, i.e., cognitive, metacognitive, and motivational, as essential components of self-regulated learning are present during all phases of learning. Prominent researchers propose that self-regulated learning can be viewed as a three-phase process: before dealing with the task at hand, during dealing with it, and after the task is completed (see Winne, 1996; Zimmerman, 1999). Similarly, strategic action of good readers as an aspect of self-regulated reading is present from the moment they are faced with a text until after completing the reading event (see Pressley & Afflerbach, 1995).

More specifically, before reading, a good reader is able to plan his activities from the beginning, the subgoals of action, the means, etc., through which he will increase the possibilities to achieve his ultimate goal. This means that good readers think and act metacognitively in advance. Once actual reading begins, skilled readers are able to distinguish important information or to skip information that is not relevant to their reading goals, to predict what is coming up next, and to analyze and combine activities and information (Gourgey, 2001). Skilled readers while reading might also activate prior knowledge, generate questions, and pay attention to confusing or inconsistent points (Pressley & Hilden, 2006). The above are basic cognitive skills that facilitate performance in many achievement situations. Good readers are also highly interpretive and evaluative of the text during reading and they do a lot of monitoring of the reading and the comprehension processes. They might need to adapt their goal, perhaps making it shorter termed and more specific or even changing it completely, or they may need to adjust their strategies (Horner & Shwery, 2002). These metacognitive skills of monitoring, evaluating, and regulating the cognitive processes are also highly valued by researchers as important facilitators of learning and performance. Furthermore, successful readers take initiatives, show high levels of self-activation, persist in the face of reading difficulties, and motivate themselves effectively throughout the activity. Recent researchers argue that such motivation and regulation of motivation activities are also a critical determinant of students' learning and achievement (Weinstein et al., 2000; Zimmerman, 1999).

When good readers make it through a text once, they evaluate themselves to confirm that they understand and remember what they have read (Horner & Shwery, 2002). When the reader senses that something is missing from his understanding, this can motivate additional reading of the text and he might decide to read more slowly, deliberately reflecting on the text. In summary, skilled reading is massively strategic, involving metacognitive processes and relating ideas of a text to prior knowledge (Pressley & Hilden, 2006). Good readers are skilled, active, and self-regulated before, during, and after reading using the repertoire of their skills and strategies to the full.

On the other hand, students' failure to control and regulate their learning and problem-solving processes and limited strategic skillfulness have been associated to poor performance and learning problems (Butler, 1998; Gourgey, 2001; Jacobs & Paris, 1987; Lovett et al., 1996; Oakhill & Cain, 2000). Several studies on good reader–poor reader differences in text processing suggest that poor readers fail to (a) conceptualize reading as a search for meaning, (b) monitor their comprehension to ensure that they are deriving meaning, (c) engage in strategic behavior to bring meaning when there has been a breakdown in comprehension, and (d) modify their choice of strategies to meet the varying demands of reading (Horner & Shwery, 2002). Furthermore, poor readers do not clarify adequately the relationships among the facts of the problem and they detect errors less often while reading in comparison to good readers (Jacobs & Paris, 1987). Poor readers tend to focus on a handful of strategies they use regardless of the particular reading situation and they have difficulties monitoring whether these strategies are working and evaluating their outcomes and the achievement of their reading goals (Gourgey, 2001). To conclude, researchers generally agree that poor readers lack engagement in strategic action and they lack monitoring skills, flexibility to adapt and regulate their reading in different situations, and control over their reading in comparison to good readers.

Rationale and Aims of the Study

Reading comprehension in a self-regulated fashion involves internal processes, such as strategic thinking, and more observable, behavioral indicators (see Zimmerman, 1999), such as verbal and nonverbal indications of strategic action. An example is self-monitoring of reading; e.g., by interrupting the reading process, examining more closely the text, and deciding to reread it. Students' overt behaviors during learning and problem-solving might be used by microgenetic methods to infer internal self-regulatory and thought processes (Siegler, 2006), such as the use of self-regulatory skills and strategies. Furthermore, recent theorists underlined the importance of investigating students' learning and strategic behavior “as an event”; i.e., during actual engagement with the task to be solved (see Winne, Jamieson-Noel, & Muis, 2002; Winne & Perry, 2000). It has been claimed that such measures may be valuable and reliable indicators of children's spontaneous problem-solving processing (Siegler, 2006; Whitebread et al., 2005). In the present study, various aspects of students' problem-solving behaviors were assessed through direct observation of what the students actually did instead through a self-report instrument. This method was chosen as the most appropriate because of the young age of the participants and because strategies are skills and this is a direct way to assess them. In the present study, students' overt problem-solving behaviors were recorded by different observers as external indicators of cognitive, metacognitive, and motivational skills, strategies and tactics that students employ to perform and regulate the reading comprehension process.

The need to find more effective ways to gauge students' strategic processing in situ in order to allow us to glimpse the inherent interplay of cognitive, motivational, and other forces during learning and achievement has been underlined (Alexander et al., 1998). In this study, different cognitive, metacognitive, and motivational aspects of students' problem-solving behaviors were assessed, aiming at investigating and comparing the profile of strategic behaviors between high and low reading comprehension achievers. The different strategic behaviors assessed in the study were identified in previous literature as self-regulatory components that serve as general or task-specific procedures that are presumably related to achievement in various cognitive situations (Zimmerman, 1999). Finally, another aim of the study was to investigate the pattern of relationships between the different strategic behaviors and subsequent performance in reading comprehension tasks in high and low achievers. According to previous literature, the main hypothesis was that high reading comprehension achievers should outperform low reading comprehension achievers in the employment of all the strategic behaviors examined during their efforts to comprehend the text.

Method

Participants and Selection Procedure

Phase a—Group Assessment in Reading Comprehension

In the first phase of the study the main aim was to select the high and low reading comprehension achievers in order for them to participate in the individual examination of the second phase. The participants of the first phase were 127 third graders (mean age: 9 years, 7 months), 71 boys and 56 girls, from five primary city schools. Third graders were chosen because they are not novice readers, as, for example, first graders, while at the same time, their reading skills and strategies are still developing. It was considered important to investigate strategic behaviors in reading comprehension in these ages where individual differences in strategy use become more evident.

Reading Comprehension Test for Groups

In the first data collection phase, a text (a folk story), selected and adapted, after a pilot administration, to the needs of the present study, was administered to all 127 participants during regular school hours. The instrument was developed and structured following the Greek language course curriculum for third graders. The participants were examined in three skills that are considered important for reading comprehension according to previous literature (see Gourgey, 2001; Jacobs & Paris, 1987; Lovett et al., 1996; Palincsar & Brown, 1984; Pressley & Hilden, 2006; Schorzman & Cheek, 2004). Those skills were prediction of the end of the story (one task), answering content questions (two tasks), and summarization (one task). The maximum score a student could obtain was 12 points. The participants' mean performance in the present reading comprehension test was 5.81 (SD = 2.46).

The students read the text individually and, afterwards, they were asked to start doing the first task. The beginning and the end of the working out of each exercise were orally indicated by the examiner. The students were given a 40-minute time limit for the whole test completion.

Overall, 45 students were selected from the initial sample to participate in the second phase of the study. Specifically, 20 high and 25 low achievers were identified by means of the above mentioned reading comprehension test for groups. There were 26 boys and 19 girls. The statistical criterion for selecting students with high and low performance in the reading comprehension test was one standard deviation above and below the mean performance in that test. Moreover, the students' teachers were asked to rate their best and their worst students in reading comprehension. Their answers generally confirmed the results by the statistical criterion we used. A few students who did not meet the two above requirements (i.e., one standard deviation above and below the mean and the teachers' nomination) were excluded from the study.

Research Instruments

Phase b—Individual Testing in Reading Comprehension

The 45 students selected for the second phase of the study were individually examined in a second reading comprehension test. This reading comprehension test was developed for the requirements of the present study, aiming at a more thorough assessment of the behaviors employed by the readers during their efforts to understand the given material. Therefore, the test procedure was structured in steps of action. Each student's efforts to comprehend the material were video-recorded.

The individual reading comprehension test consisted of a two-page passage that referred to a mystery story entitled “The Stolen Treasure: A Mystery Story.” Pilot-testing allowed us to better adapt the text to the needs of the study. The reading comprehension test was built on the basis of previous literature that identified specific skills and strategies as important for reading comprehension (see Gourgey, 2001; Q4 Jacobs & Paris, 1987; Palincsar & Brown, 1984; Paris & Paris, 2001; Pressley & Hilden, 2006; Schorzman & Cheek, 2004). For example, important attributes of the skilled readers described in literature that are frequently included in instructional programs for reading comprehension are to make predictions about the text on the basis of the title and on looking at headings and pictures, to be able to summarize large sections, to find the main idea, to attend the sequence of important events in a story, etc. (Gourgey, 2001; Pressley & Hilden, 2006). Paris and Paris (2001) reported that key strategies in reading are to make inferences, to answer content questions, to elaborate the meaning from the text, and to identify main ideas. On the basis of this literature, the participant students were examined in the following tasks: predicting the context of the text by its title (one task), answering content questions (three tasks), putting sentences in the correct order (one task), recognizing a hero in a given photo according to his description in the text (one task), putting the words of a sentence in their correct order (one task), marking on a given map an itinerary (two tasks), finding the timetable of a journey (one task), pinpointing the key words in the text, and, finally, finding out the main idea of the text (one task). The maximum score that one could obtain by summing up performance in the above tasks was 26 points. For the purposes of the present study, only quantitative data were taken into account.

Structured Observation form for Strategic Behavior

A structured observation form was used to assess students' strategic behavior during reading comprehension. This is an instrument that includes the assessment of different behaviors as indicative of students' thinking and employment of problem-solving strategies. Most of the strategic behaviors assessed were proposed by Zimmerman (1999) as presumably related to the students' achievements in various cognitive circumstances (Zimmerman, 1999; see also Hwang, 1998; Palincsar & Brown, 1984). The instrument also involves assessment of students' retrospective evaluations of the learning outcome and initially it was based on Hwang's work (1998).

Eleven strategic behaviors employed by the students to perform and regulate the tasks at hand were assessed (see Appendix for their top-ranking criteria). The behaviors tapped cognitive aspects of strategic behavior (behaviors 1–3, α = .92), metacognitive aspects (behaviors 4–7, α = .97), and regulation of motivation (behaviors 8–11, α = .95), following a frequent theoretical taxonomy proposed in literature (see Gourgey, 2001; Kuyper et al., 2000). We did not make use of this taxonomy in the statistical analyses of our data because the limited number of participants did not allow for testing its structural validity. The observation form's internal consistency and structural validity have been confirmed in previous studies (Dermitzaki, 2005; Dermitzaki, Leondari, & Goudas, 2004, in press).

Two previously trained judges observed students' video-recorded efforts and recorded each student's employment of the aforementioned behaviors based on specified criteria of assessment. In general, the scoring criteria corresponded to the frequency with which the student employed each behavior. This assessment was based on a 4-point scale (for the top-ranking criteria see Appendix; for a full description of the criteria used, see Dermitzaki, 2005). Each judge gave 132 estimations for each student (11 behaviors × 12 tasks). The two judges rated each behavior independently, and when there were large differences in their assessment, there was a discussion between them and a third rater to reach the greatest degree of agreement. The two judges' ratings for each student were summed up and divided by the number of the assessed behaviors. The final score of each student's use of strategic behaviors was the mean of the two judges' ratings.

Cross-tabulations and the gamma statistic for each behavior were calculated in order to measure the degree of agreement between the two judges' ratings. The mean of the gamma values for the 11 behaviors was .93, indicating a high degree of agreement between the two observers.

Procedure

The 45 students selected to participate in further examination in reading comprehension were examined individually during regular school hours in a quiet room of the school. The examiner explained to the student the procedure they were going to follow with the two-page passage. Then, the student was asked to read aloud the title of the given passage and to predict its content. Next, the student was asked to read the first part (first page) of the passage. After dealing with the tasks corresponding to the first part, the student was asked to read the second part of the passage and perform the rest tasks. No further prompts were given to the students. Each individual student's efforts to comprehend the given text were video-recorded. The individual examination procedure lasted approximately 40 minutes.

Results

The means and the standard deviations for the reading comprehension test and for each behavior employed by the students are shown in Table 1 separately for the total sample and for the high and low achievers. Table 1 also shows the results from a multiple analysis of variance and the effect sizes (index η²) with independent variable the groups of the high and low achievers and dependent variables the students' performance in the reading comprehension test and the 11 strategic behaviors assessed.

TABLE 1 Descriptive Statistics and Differences Between High and Low Reading Comprehension Achievers

	Total Sample		High Achievers		Low Achievers
	M	SD	M	SD	M	SD	F	df	p	ES
Comprehension	15.08	5.11	19.70	2.51	11.40	3.31	85.669	1, 43	.000
achievement
Concentration	3.41	.71	3.86	.19	3.05	.77	21.242	1, 43	.000	.33
Analyzing/combining	2.55	.94	3.46	.44	1.82	.49	135.496	1, 43	.000	.76
activities
Choosing main	2.56	.90	3.46	.40	1.85	.43	164.884	1, 43	.000	.79
and trivial
Planning	2.52	1.02	3.52	.46	1.72	.51	150.291	1, 43	.000	.78
Monitoring	2.57	1.00	3.52	.44	1.81	.57	120.326	1, 43	.000	.74
Evaluating	2.45	.87	3.26	.54	1.80	.41	104.872	1, 43	.000	.71
Awareness of	2.60	.91	3.48	.44	1.89	.45	137.907	1, 43	.000	.76
errors
Initiative	3.25	.67	3.70	.33	2.88	.66	25.575	1, 43	.000	.37
Working	3.29	.61	3.71	.28	2.95	.59	27.812	1, 43	.000	.39
autonomously
Persistence	3.42	.63	3.88	.19	3.06	.63	31.097	1, 43	.000	.42
Maintaining	3.42	.62	3.85	.25	3.15	.66	19.913	1, 43	.000	.32
motivation

In general, the means denote moderate to high employment of the strategic behaviors by the 45 participant students. Concerning the high and low achievers' performance in reading comprehension tasks and their use of strategic behaviors, as expected, the MANOVA showed statistically significant predominance of the high achievers over the low achievers regarding their performance and their use of the strategic behaviors during text comprehension (F _{(12, 32)} = 16.403, p = .000). More specifically, while the high achievers were generally found to extensively employ all the strategic behaviors, the low achievers were rated extremely low regarding their employment of metacognitive and cognitive components of strategic behavior. Inspection of the effect sizes in Table 1 shows that relatively large differences between high and low achievers in reading comprehension were found in the metacognitive aspects of the behaviors assessed, such as planning, awareness of errors, and adjustment of aims, and in the deep cognitive processing aspects; i.e., in choosing between main and trivial information and in analyzing and combining activities. The rest of the behaviors assessed presented small to moderate differences between the two groups.

In order to investigate the relations between the students' reading comprehension achievement and their actual strategic behaviors, a Pearson correlation matrix was computed. Table 2 presents these correlations separately for the sample of 45 students and for the high and low achievers.

TABLE 2 Pearson Correlations Between Performance in Reading Comprehension and Employment of Strategic Behaviors in High and Low Reading Comprehension Achievers

	Reading Comprehension Performance
	Total	High	Low
	Sample	Achievers	Achievers
Concentration	.71∗∗	.58∗∗	.55∗∗
Analyzing and combining activities	.93∗∗	.80∗∗	.77∗∗
Choosing between main and trivial	.93∗∗	.77∗∗	.78∗∗
Planning	.94∗∗	.84∗∗	.81∗∗
Monitoring of the activities	.93∗∗	.76∗∗	.77∗∗
Evaluating	.91∗∗	.74∗∗	.77∗∗
Awareness of errors	.93∗∗	.76∗∗	.83∗∗
Initiative	.72∗∗	.46∗	.51∗∗
Working autonomously	.71∗∗	.24	.50∗∗
Persistence	.77∗∗	.57∗∗	.59∗∗
Maintaining motivation	.72∗∗	.74∗∗	.53∗∗
Notes.

Table 2 shows that all of the behaviors employed by the 45 participant students were significantly related to their subsequent performance in reading comprehension. The metacognitive and cognitive aspects had the strongest correlations to the students' performance (e.g., planning, monitoring of the learning process, analyzing and combining activities, and choosing between main and trivial). This pattern of significant relationships among achievement in reading comprehension and strategic behaviors was generally maintained in the Pearson correlation matrices for both the high and low achievers.

Discussion

One aim of this study was to investigate the profile of strategic behaviors during reading comprehension in high and low achievers. The results of the study showed that, in general, third graders were relatively able to use a repertoire of skilled and strategic behaviors during their efforts to understand the given text, corroborating recent research that documents that students can be strategic at least to a degree early in their school life (Juliebö et al., 1998; Perry, 1998; Siegler, 2000; Whitebread et al., 2005). More specifically, the participants were found to sufficiently regulate their motivation toward the tasks at hand, since they were found to persist in the face of difficulties, to retain their interest in the activities, and to remain concentrated on them. However, the students displayed moderate employment of metacognitive and cognitive strategic behaviors to perform and regulate their reading comprehension process. Thus, our participants at the age of 9–10 years seem efficient to regulate their motives and persist in their efforts but may still have difficulties in cognitively and metacognitively performing and regulating adequately the comprehension process. This finding agrees with previous research that claims that younger students' metacognitive skillfulness is still developing; therefore, they use fewer metacognitive strategies (Butler, 1998; Veenman & Elshout, 1999). It is suggested that, even by grade 6, many of the students have difficulties in identifying main ideas and distinguishing important from unimportant information, and they do not understand well what they read and they often fail to monitor and repair the problem-solving process (Paris & Paris, 2001; Pressley, 2002; Pressley & Hilden, 2006). It is also known that even high school graduates and beginning college students are not metacognitively mature with respect to reading (Pressley & Afflerbach, 1995). Although children at about 9–10 years of age have rudimentary understanding of the task of reading, they have incomplete concepts about the nature and purpose of reading, and their understanding and control of strategies, especially metacognitive ones (Juliebö et al., 1998; Paris, Wasik, & Turner, 1991), are often inadequate and will increase with skill and age.

The findings of the study also showed that high achievers outperformed low achievers in employing all the strategic behaviors assessed, thus corroborating previous research (Brown & Pressley, 1994; Horner & Shwery, 2002; Jacobs & Paris, 1987; Stahl, 2004). Furthermore, this study showed that the two groups had a relatively different profile in their strategic behaviors. More specifically, it was revealed that high achievers effectively regulated both their cognition and their motivation toward the tasks to be solved. Low achievers, on the other hand, although they regulated adequately their motivation to the tasks at hand, they insufficiently employed the metacognitive and cognitive strategic behaviors to perform and regulate their efforts to comprehend the material. This finding is in line with previous literature on good reader–poor reader differences in text processing (Horner & Shwery, 2002; Palinscar & Brown, 1984; Pressley & Hilden, 2006). Previous studies suggested that poor readers fail to engage in strategic behaviors, which implies either no use or ineffective use of cognitive and metacognitive strategies. Therefore, they may use strategies that are not appropriate to the task they have to deal with and persist in using them even if they are not effective for a particular situation.

More specifically, relatively large differences between the high and low achievers were detected regarding the behaviors, indicating distinction between important and unimportant information, their practices to analyze and combine activities, and in all the metacognitive-in-nature behaviors indicating planning, awareness and repair of errors, monitoring of efforts toward reaching a solution, and evaluation of the outcome. Past research has shown that, as the good readers proceed through the text, they identify important information and are able to analyze and combine activities (Gourgey, 2001; Pressley & Afflerbach, 1995). Moreover, high achievers displayed, to a large extent, planning behaviors, whereas on the part of low achievers, planning behaviors were displayed the least of all. Planning is a prereading activity referring to the selective coordination of a cognitive means to a cognitive goal (Jacobs & Paris, 1987). It helps readers set their goals in reading and understanding a text. For example, a student decides, first, to carefully read the text; second, to find out its main idea; third, to group meanings into coherent units; and so on. Previous findings showed that inadequate reading behaviors starts from the prereading stage during which poor readers are not adequately prepared in order to proceed with the reading process itself (Williams, 1998). Our results supported such reports. As mentioned above, relatively large differences between the two groups, in favor of the high achievers, were also traced regarding the students' awareness of errors and subsequent adjustment of intermediate aims. For example, after the student has put the sentences given in the order he thinks correct, he finds out that there is no logical meaning extracted when reading them, and he is trying to find out the correct order. It has been suggested that good comprehenders perceive errors more often in comparison to poor comprehenders while reading (Jacobs & Paris, 1987) and that better learning occurs through increasingly precise fitting of strategy choices to the demands of problems and situations (Siegler, 2000). Thus, what differentiated the high from the low achievers was mainly the employment of deep cognitive and metacognitive strategic behaviors.

Strong relationships were detected among students' strategic behaviors during task engagement and their reading comprehension achievement, thus confirming previous findings (Gourgey, 2001; Q4 Pressley, 2002). These strong correlations revealed might be in part due to the assessment of the variables of interest in a task specific level. All of the 11 behaviors were positively and highly related to the students' performance in comprehension tasks. Previous findings showed that strategic behavior is clearly related to students' academic achievement and that there are great benefits in reading comprehension from comprehension strategies instruction programs (Brown et al., 1983; Gourgey, 2001; Pintrich & DeGroot, 1990; Pressley, 2002; Pressley & Hilden, 2006; Veenman & Elshout, 1999). Future longitudinal research designs should specify how the relationships between employment of skills and strategies, students' initial academic achievements, and their effects on subsequent self-regulatory use of strategies are shaped in the long run.

The pattern of the strong relationships found between task performance and employment of mainly metacognitive-and cognitive-in-nature behaviors was maintained both for high and low achievers. Application of behaviors for regulating motivation during efforts to comprehend the material had moderate correlations to respective performance for both groups. These findings again underline the major contribution of cognitive and metacognitive processes in problem-solving situations (Brown et al., 1983; Flavell, 1979; Pressley & Hilden, 2006).

Overall, the above findings suggest that being able to identify students' strategic strengths and weaknesses during performing the task at hand enables teachers to choose the most appropriate instructional methods and practices for a particular student and the particular type of task. Microgenetic analyses have consistently indicated that quantitative changes in execution of strategies might play a large role in learning (Siegler, 2006). Moreover, early identification of young and poor readers' difficulties in applying text comprehension strategies would be useful for subsequent intervention in order to enable them to become autonomous and engaged, self-regulated readers. Teachers and educators need to increase the youngest and the poor comprehenders' awareness of the task demands, instruct them as to which are the appropriate skills and strategies to employ in order to facilitate task completion, and teach them how to monitor and regulate the application of these strategies in different situations.

Some remarks regarding the design and the methodology of the study should be added. The findings of this study should be interpreted with caution, since they concerned a limited number of students belonging to a particular age group examined using a text with a specific content and structure and in a limited number of comprehension tasks. The limited number of participants also did not allow us to document thoroughly the reliability and validity of all the assessment instruments. The aims of the present study led us to develop our own instruments since it was not easy to find appropriate instruments to assess reading comprehension and use of strategies in a non-English language. Therefore, this study is best described as a pilot study in the domain of reading comprehension. Further research is needed with larger samples of various age groups and using a variety of reading comprehension tasks. It is also important that, in future studies, educational context variables, text variables, and student variables should be taken into account, such as texts with various structures and information level, students' decoding and memory skills, prior knowledge, and level of verbal and general intelligence. The role and the contribution of each particular strategic behavior in respective performance are also issues that need to be addressed in the future. Future research should further develop instruments for assessing actual strategic behaviors and profit from the match between self-report measures and measures of students' employment of strategic behavior “in action” (Juliebö et al., 1998; Winne et al., 2002; Zeidner, Boekaerts, & Pintrich, 2000). Such data will offer valuable information regarding the on-line processes of self-regulated learning because, as Siegler (2006) recently noticed: “The only way to find out how children learn is to study them closely while they are learning” (p. 469).

Appendix

The Top-Ranking Criteria for Observers on Each of the Behaviors Assessed

1. Concentration—Perceives external stimuli but is not distracted by them

2. Analyzing and combining activities—Joining small parts resulting from previous activity to make a meaningful whole

3. Choosing between main and trivial—Methodically selects the substantial elements, ignores the trivial ones

4. Planning—Working with a clear plan, using time effectively

5. Monitoring of the activities—Examines closely the solution process, selects appropriate next step

6. Evaluating (in the discussion after the solution)—Offers evaluations after observing the outcome

7. Awareness of errors, adjusting intermediate aims—Is aware of errors and tries to correct them

8. Initiative (starts action on his own)—Shows initiative and high levels of self-activation, decides next step with no need for intervention

9. Working autonomously—Works autonomously, needs no intervention or reinforcement by the experimenter

10. Persistence—Works persistently in face of difficulties till finding a solution

11. Maintaining motivation—Effectively motivates himself and retains interest for the activity

Notes

∗p < .05,

∗∗p < .01.