The Good Strategy User Model
Table of Contents:
This model describes three categories of strategies: goal- or task-specific strategies, monitoring strategies used to control and regulate goal-specific strategies, and higher-order strategies used to plan sequences of goal-specific and monitoring strategies (Pressley, 1986; Pressley, Borkowski, & Schneider, 1987). While the monitoring and planning strategies are considered forms of metacognitive procedures, metacogni-tive knowledge is represented in the model as specific strategy knowledge, including knowledge about how, when, and where to use particular strategies. Such knowledge also has a motivational aspect because it includes understanding that success is due to the use of task-appropriate strategies and that failure might have been avoided by the use of such strategies. In addition, general metacognitive knowledge about strategies has motivational properties because it includes understanding that strategic effort generally increases the likelihood of success. Finally, the good strategy user has an extensive knowledge base (in addition to the knowledge of and about strategies) that sometimes may make strategic effort superfluous and sometimes prompt strategy use but, most importantly, enables use of particular strategies (Pressley, 1986; Pressley et al., 1987).
Perhaps the most controversial aspect of the Good Strategy User Model is that it posits that good strategy use is often characterized by automaticity, implying that the components described above, including strategic processing, have been automatized or habituated and therefore do not require conscious attention. As noted above, such automatic processing tends to be categorized as skills rather than strategies in more contemporary theorizing.
Because the research leading up to the Good Strategy User Model overwhelmingly relied on randomized experiments, in which participants were trained to perform particular strategies and the effects on outcome variables such as recall of information were measured, the measurement of strategies per se was not a major issue. However, as noted by Levin (2008), Pressley was also eager to use qualitative methods, such as interviews or cued recall techniques, to gain insights into participants’ thinking about their cognitive processing (e.g., Pressley & Levin, 1977), and he later became a strong proponent of verbal protocol analysis (e.g., Pressley & Hilden, 2004).
The Model of Domain Learning
This model describes the interplay of knowledge, strategies, and interest across three stages of academic development within a domain, which are termed acclimation, competence, and proficiency/expertise (Alexander, 1997, 2004, 2005, 2012). In addition to changes in the configuration of these main components that occur across the stages, the model acknowledges that there are many phases or episodes of learning within each stage, with these phases or episodes also characterized by certain interplays among knowledge, strategic processing, and interest (Alexander, 1997).
Regarding strategic processing, strategies at different levels of specificity (i.e., domain-general vs. domain-specific) and processing (i.e., surface- vs. deep-processing) are distinguished, as well as cognitive and metacognitive strategies (Alexander, 2004, 2005). Importantly, strategies are conceived of as effortful, intentional, and purposeful procedures directed toward improving learning, comprehension, and problem solving through the acquisition, transformation, and transfer of information (Alexander, 1997, 2004). While learners’ dependency on surface strategies to establish a rudimentary knowledge base, gain foundational understanding, and solve elementary problems decreases during the acclimation stage, essentially levels off during the competence stage, and then again decreases during the proficiency/expertise stage, their use of deeper strategies to organize, transform, and critically analyze information increases across the three stages of domain learning and becomes particularly important in the service of knowledge generation, deep comprehension, and problem formulation during the proficiency/expertise stage (Alexander, 2004, 2012).
Of note is that combined shifts in knowledge and interest across the stages are seen as important contributors to the development and increased use of deeper processing strategies (Alexander, 1997). In effect, mutually influential relationships among a large, well-integrated body of knowledge, high individual interest, and a well-established and efficient repertoire of deeper processing strategies are regarded as a hallmark of the most advanced learners in a domain (Alexander, 1997, 2004).
With respect to the measurement of strategies, Dinsmore, Hattan, and List (2018) documented that the research conducted within the Model of Domain Learning overwhelmingly has measured strategic processing by means of offline self-report inventories. However, these offline self-reports have typically been task-specific and collected immediately after task completion (Alexander & Murphy, 1998; Alexander, Murphy, Wood, Duhon, & Parker, 1997; Alexander, Sperl, Buehl, Fives, & Chiu, 2004). For example, Alexander et al. (1997) asked participants to monitor the strategies they used during task completion (i.e., text reading) and immediately afterwards check any strategy they had used to comprehend and remember the text on a list of 20 text-processing strategies, also marking the strategies they had found most helpful. As Dinsmore et al. (2018) showed, the reliability estimates reported for the kind of strategy measures used within the Model of Domain Learning have quite often been lower than desirable, and the relationship between scores on such measures and performance actually seems open to question.
The Cyclical Model of Self-Regulated Learning
This model describes the three cyclical phases of forethought, performance, and selfreflection (Zimmerman (2000, 2013). Forethought includes task analysis and selfmotivation used in preparation for efforts to learn. The performance phase involves the execution and monitoring of strategies planned during the forethought phase. Finally, during self-reflection, individuals self-evaluate their learning and reflect on the causes of the outcome, as well as react emotionally to what happened during performance and draw inferences regarding future learning. The cyclical nature of the model involves that processes in one phase influence processes in the next, and that processes during self-reflection influence processes in the forethought phase when individuals continue their efforts to learn (Zimmerman, 2000, 2013).
Within Zimmermans model, self-regulated strategies are conceived of as purposefully selected or planful cognitive processes and behavioral actions directed at acquiring or displaying knowledge and skills. For example, strategies can facilitate learning and performance by helping students attend to, analyze, and reorganize academic tasks
(Zimmerman, 2000). Of note is also that strategies are considered context-specific within self-regulated learning theory, implying that self-regulated students adjust their strategic choices and activities to different study contexts (Zimmerman, 2000). Finally, Zimmermans view on self-regulated learning strongly emphasizes that strategic competence is of little value if individuals cannot motivate themselves to use it, with one key source of motivation being their self-efficacy perceptions.
Zimmermans research on the identification and measurement of self-regulated strategies is strongly associated with the Self-Regulated Learning Interview Schedule (Zimmerman & Martinez-Pons, 1986, 1988, 1990). This methodology consists of a 15-minute individual structured interview during which students are presented with different hypothetical learning contexts (e.g., when completing writing assignments outside class). For each context, students are asked to describe the methods they would use, and if they mention one or more strategies for a learning context, they are also asked to rate the frequency with which each mentioned strategy is used. Zimmerman and Martinez-Pons (1986, 1988, 1990)) have shown substantial positive correlations between students’ reports of strategies on this schedule and their academic achievement. However, in addition to using such offline self-reports about hypothetical learning contexts, Zimmerman and colleagues (Cleary & Zimmerman, 2001; Kitsantas & Zimmerman, 2002) have more directly evaluated processes included in the cyclical model by asking individuals close- and open-ended questions during actual task completion. We will return to this methodology in the section termed Task-specific Self-report Inventories.
To summarize, the three reviewed models provide a foundational understanding of the importance and functioning of strategic processing. Taken together, they describe how strategic processing, functioning within a system of cognitive, metacognitive, and motivational components, can improve learning, comprehension, or problem solving. The models do, however, differ with respect to their adoption of a developmental perspective and the way they conceptualize strategic processing. Thus, while Pressley (1987; Pressley et al., 1987) certainly did not disregard the importance of development, the Good Strategy User Model is not a developmental model of strategic processing. By comparison, both the Model of Domain Learning and the Cyclical Model of Self-Regulated Learning could be described as developmental. These two models differ in terms of the grain sizes on which they focus, however, with the former focusing on comprehensive stages of academic development and the latter focusing on more fine-grained phases that unfold sequentially in a recursive manner over continued efforts to learn. With respect to the conceptualization of strategic processing, Pressley’s model differs from the others because it allows for automatic strategic processing, and Alexander’s model is unique in distinguishing between surface- and deep-processing strategies. Finally, we note that research conducted within the three models has largely relied on self-reports of strategic processing. In the following section, we further elaborate upon some of these self-report methodologies, ranging from thinking aloud during task completion to answering questions about task-specific processing retrospectively.