Nurturing self-regulated learners: ramifications and reflections

Mrs Seah-Tay Hui Yong Nanyang Girls' High School Singapore

Introduction

Lately, many countries have found cause to be concerned about the outcomes of their educational systems. For example, the Australian Council of Deans of Education (2001) concluded after an extensive study that the Australian education is ‘at best uncompetitive, and arguably in a state of crisis’ (p. 1). In the 2003 Trends in International Mathematics and Science Study, not only were the American students outperformed by their Asian peers, they also performed worse than in 1995 (Gonzales et al. 2004).
Even in Singapore, which emerged tops in the study, it appears that the system produced workers who lack ‘creativity and initiative’, who are unable to ‘think laterally’ ('Learning, Creating, Communicating: A Curriculum Review', 1998, p.4). This remark was by the External Review Team (ERT), who then suggested that this was due to the ‘repetitive, structured exercises using workbooks and assessment sheets to train them to produce the right responses to examination questions’ (p. 7), which has led to a lack of the ‘spirit of inquiry’ in our primary students and ‘a lack of responsibility for their own learning’ in the secondary students (p. 12). Their report subsequently led to achieve the Teach Less, Learn More (TLLM) initiative, which exhorts teachers to ‘encourage more active and engaged learning in our students’ and to ‘motivate students to take ownership of their own learning’ (Teach Less, Learn More, 2007).

Many educators will applaud this call to develop self-regulated learners. Advocates like Zimmerman (2001) argue that a fundamental goal of education is guiding one’s learning in school — but also beyond (Boekaerts, 1997).

There are various definitions of self-regulated learning but there are core features that correspond to the oft-quoted definition by Schunk and Zimmerman (1994): ‘self-generated thoughts, feelings, and actions which are systematically oriented toward attainment of their goals’. One feature of self-regulated learning is that students are assumed to be aware of the potential usefulness of self-regulatory strategies, such as planning to manage and control their effort on classroom academic tasks. This stems from the constructivist view that students, rather than being passive recipients of information, contribute actively to their learning goals and exercise control over goal attainment by actively seeking and processing information (Schunk, 2001). Put succinctly, learning is not something that happens to students but rather by students (Zimmerman, 2001).

The second feature is a self-oriented feedback loop in which students monitor the effectiveness of their learning methods or strategies and respond to feedback in a variety of ways. However, such knowledge of strategies is a necessary but not sufficient condition for student achievement. To be successful, students must have both the ‘will’ and the ‘skill’ (Pintrich & DeGroot, 1990). Boekaerts (2002) argues that students bring their own goals to the classroom and so self-regulated learning should rightly address the whole-person-in-context by acknowledging the students’ socio-emotional goals.

While the goal of developing self-regulated learning in our students is intuitively appealing, what are its ramifications? For a start, the new desired learner outcomes will conceivably place very different demands on the teacher, compared to the days of teach and drill. And if students are not merely passive participants in class, what demands will be made of them, as well?

While it is tempting to provide answers and prescribe solutions, this paper aims to raise and discuss these issues in the light of a study I conducted recently.

Background to study

Researchers have long known that equipping students with the skills to self-monitor, make judgments of their own learning and use feedback for their own performance can improve student performance. For example, the 1996 meta-study by Hattie, Biggs, and Purdie recommended strategy training, where learners participate actively in self-assessment, taking responsibility to decide the strategies that that are appropriate to the task at hand. Similarly, Swanson’s (2001) meta-analysis of intervention research found significant effect size, especially on measures of higher-order processing (for example, planning, self-questioning) and text-understanding (for example, inferential comprehension, thematic understanding). Flavell, who coined the term ‘metacognition’to describe such knowledge and thinking processes associated with the use of strategies, said that ‘good schools should be hotbeds of metacognitive development’ (1987), with children having repeated opportunities to monitor and regulate their cognition, as metacognition improves with practice.

Others have agreed that students cannot become self-regulated learners without a curriculum that allows it to happen. Schools should, for example, give opportunities for student choice in the way mastery of content and subject matter is demonstrated (Northwest Regional Educational Laboratory, 2004). How else can students develop the ability to contribute actively to their learning goals and exercise control over goal attainment by actively seeking and processing information (Schunk, 2001)?

Another recommendation includes problem-based learning. A study by Palmer and Wehmeyer (2003) seems to support this as they concluded that problem-solving and goal-setting have been found to help develop self-regulation in students as young as aged five to six.

Specifically, these problems for exploration should be sited in the real world. Social cognitive theorists, in particular, emphasise the importance of the situational context of the learner as ‘student responses are coloured by the particular task and setting’ (Wiggins, 1993). They believe that embedding relevant, real-life support in classroom learning facilitates the emergence of self-regulated learning (Borkowski, Weaver, Smith, & Akai, 2004). Their claim of the benefits of such situated learning is supported by the meta-study by Hattie et al. (1996).

My own study aimed to verify whether a real-world context facilitates self-regulated learning, by comparing the students’ use of self-regulated learning strategies in two contexts: a conventional paper-pen task and an authentic task. The former involved a writing task called Stimulus Response Question (SRQ), in which students are asked to critique a piece of exposition by evaluating the writer’s arguments, as well as providing their own insight into the issues raised. The authentic task was writing a letter to the editor of one of the local newspapers in response to a news article or a letter published on the forum pages. In essence, this second task draws upon the same SRQ skills but in the context of an authentic audience (the editor and the public, if the letter is published). For the purpose of studying any possible effects of peer support on SRL, I had given them the choice of writing the letter in a group, while the SRQ was individual work.  

After the SRQ task, I asked the classes to fill in a survey. The 20 questions in this Likert-scale survey were based on suggested indicators of self-regulated learning (SRL) strategies by Zimmerman and Martinez-Pons (1986) and Zimmerman (2002). The questions are as follows, with associated SRL strategies indicated within brackets:

• I checked my work while I was doing it (self-evaluation).
• I made an outline before I wrote my answer (organising).
• I set aside sufficient time to attempt this assignment (planning).
• Before beginning to write my answer, I tried to get as much information as possible concerning the topic (seeking information).
• I kept my previous SRQ answers / models of SRQ answers (keeping records).
• While writing my answer, I isolated myself from anything that distracted me (environmental structuring).
• I would give myself a treat if I did well with this assignment (self-consequences).
• I have practised the steps needed in SRQ (rehearsing).
• If I have problems with SRQ, I ask my peers for help (seeking social assistance).
• If I have problems with SRQ, I ask my teachers for help (seeking social assistance)           
• If I have problems with SRQ, I ask adults for help (seeking social assistance).
• While and/or before writing, I reviewed my notes on SRQ (reviewing records).
• I attempted to discover the main ideas in the passage. (strategic planning).
• When necessary, I changed my technique or strategies. (task strategies).
• I think I can predict quite accurately my grade for this assignment (self-evaluation).
• I think I have the ability to do this assignment satisfactorily. (self-efficacy).
• I was aware of directing myself through the various steps needed in writing the SRQ answer (self-instruction).
• I find SRQ interesting (intrinsic interest).
• I am satisfied with my answer (self-satisfaction).
• If I do well in this assignment, it will be because the strategies I used (causal attribution).

After the letter to the editor task, I administered essentially the same survey again but with references to SRQ amended to reflect the letter task. The intention was to compare the students’ responses between the two tasks to see if the nature of learning tasks has a bearing on SRL, as argued by social cognitivists (Zimmerman & Martinez-Pons, 1988).

Findings of study and discussion

While the aggregated mean score for the letter was higher (indicating greater use of SRL strategies), the difference was not statistically significant from than for the SRQ. The reason lies in that only two paired questions yielded significant difference between the two tasks: question 4 (seeking information) and question 16 (self-efficacy). Hence, it is not surprising that the aggregated scores were not significantly different. Then one would expect each respondent’s SRL scores for the two tasks to be closely related, assuming that a person’s SRL is a stable attribute. Yet, there was no significant correlation between the SRQ and letter scores for the same question. If it is that the instrument yields valid and reliable information about the respondents’ SRL strategies, then it appears that SRL is not a stable trait but is context-sensitive.

A more detailed examination of the contextual factors is made possible by looking at the data collected from interviewing a pair of students from each of the two classes involved in the study. They were chosen for having either the highest positive difference in her scores between the SRQ and the letter surveys (indicating a higher SRL score in SRQ), the other the highest negative difference (indicating a higher SRL score in letter).

The interviews cohered with the findings from the self-report survey. There appeared to be no discernible difference in the approach the participants took to writing assignments in general, including SRQ and the letter.Though they mentioned they enjoyed the company of their friends while working collaboratively on the letter, their friends’ presence did not drastically change their SRL strategies. This finding is triangulated by the survey results which saw no significant difference in question 9 (asking peers for help).

If there was any difference in approach, it was due to the requirements associated with the task. Three of the four interviewed mentioned the 100-word limit set for the letter to the editor. Hence, they were forced to ‘get to the point’ and ‘express (them)selves with more concise and succinct language’. One of them also felt that the time limit made a difference. While she had opportunities to redraft SRQ answers, she had worked on the letter in class over the two periods I had allotted to the task.

Generally, the respondents found more motivation in the letter task as they found it ‘quite refreshing’ - it felt ‘more real’, more ‘live’. One girl found the chance to apply the SRQ skills to ‘something that actually might be published’ and hence they ‘might actually be heard’, was ‘good incentives (sic) … to produce higher quality works (sic)’.

Yet another girl preferred the letter as she found she could relate to it better, as she would be able to express her views on the topic, in contrast to the SRQ task, where she was required to evaluate someone else’s views. She felt less confident about the latter task because of the many demands to pick out ‘any contradictions, loopholes, or missing premises’.  This could explain why there was no significant correlation in SRL scores across the two tasks for questions 15 (predict grade accurately) and 19 (satisfied with answer). The means for these questions are both higher in letter, though not significantly so. It could be that most students felt less certain about their SRQ fulfilling the teacher-set criteria of what makes a good answer, while they perceived themselves more in control in the letter task.

On the other hand, when I asked the students who were in my remedial class which task they preferred, they chose the SRQ, as they felt that the rubrics gave them a guide their performance. They found the fewer guidelines given for the letter task made it more daunting.

This view, that the tasks were perceived as different, surprised me as I had thought they were similar, except that one offered an authentic audience. I had overlooked how much effect the word and time limit of the letter task would have on directing the students’ performance. In hindsight, the fact that the SRQ would be graded (but not the letter) probably also contributed to the students’ relative lack of self-efficacy in this task.

A related issue is the source of the students’ sense of efficacy in both tasks. Their previous exercises and the SRQ grading rubric may help students be a better judge of their performance. In contrast, in the letter task, they had no similar previous practice nor explicit standards (though it was implied that SRQ rubric applied). Hence, a critic may question how accurately the students could assess their efficacy in the letter task.

Implications

Certainly, the findings of my study are tentative, limited by the context of two tasks and the relatively small sample size of 44 respondents, who are high ability students hailing from an elite all-girls’ school. The findings are also limited by the oft-mentioned reliability and validity issues associated with self-report instruments, for example, what respondents report may not cohere with what they do. In addition, the findings threw up some unforeseen extraneous factors. But it was these extraneous factors that serendipitously point to some tensions in efforts to develop SRL in students.

The evolving role of teacher in crafting 21st century tasks

The literature suggests that if our aim is to develop self-regulated learners, then we must create opportunities in the learning context for students to exercise such self-directedness. This implies that teachers must be skilled in designing tasks that offer such opportunities

This study showed me that, even as a senior teacher with 22 years under the belt, I was not able to anticipate some of the consequences that arose from the tasks I had designed, though, in hindsight, they were to be expected. Previous research has shown that the learner’s use of SRL is determined by the context of the task. The context itself is in a state of dynamic flux: each constituent in the context is shaped by, and in turn, shapes other constituents. In the letter task I had set, the authentic setting which I thought would give more opportunities to exercise SRL was undermined by the time and word limit.

My experience raises the question: much has been said about the skills that students will need for the 21st century, - but what about the skills needed by teachers to produce such student outcomes? To prepare them to enter a world where school knowledge becomes rapidly obsolete, teachers can no longer be mere ‘technicians’ or knowledge transmitters (Towndrow, 2004). Instead, teachers must design open learning tasks that are necessarily more ill-structured, to give students the opportunities to ‘navigate change and diversity, learn-as-they-go, solve problems, collaborate, and be flexible and creative’ (Australian Council of Deans of Education). Are teachers adequately prepared to take on this role when they themselves are products of a much more structured and predictable system? If not, then is the call for changes in the way teachers teach matched by changes in teacher training and professional development?

The evolving role of student as assessor

Another issue that arose in my study was the ability of students to self-assess. To be able to self-direct their own learning in the long run, they necessarily have to learn how to identify appropriate standards in order to make judgements about the quality of their own work. This will mean that assessment will move from the exclusive domain of assessors into the hands of learners (Boud, 2000). The implications of this are far-reaching.

Firstly, teachers lack the ‘assessment literacy’ (Stiggins, 1999) themselves. In Singapore, the problem of teachers’ lack of preservice training in assessment is compounded by being dependent for so long on centrally mandated curriculum and external examinations (PSLE, GCE ‘O’ and ‘A’ levels). The standards with which teachers judge student work probably do not deviate much from that of these external examinations. How, then, would they be able to mentor their learners in less structured contexts, where the latter have to identify appropriate standards for themselves and seek forms of feedback from their environments?

And even if they can, it is not immediately obvious that teachers want to. Students’ ability to self-direct and self-assess may empower them in a way that teachers may be uncomfortable with. After all, it is unfortunate but not uncommon for teachers to use assessments to reward or punish learners.

This power relation can also have a negative effect on students. They may have the autonomy to self-judge but they may still give in to inward tendencies to self-judge according to the teacher’s preferences (Tan, 2004).

Conclusion

Hoban (2000) argues that educational change can be compared to a spider’s web to illustrate the connectedness of the different elements involved. A change in one element will result in repercussions elsewhere in the system.

This suggests that educators and policy makers should be mindful of this in calling for a change in desired learner outcomes.

Few will dispute the merits of nurturing self-regulated learners but, as my study has demonstrated, teachers may lack the skills or experience to design ill-structured tasks that will give students opportunities to exercise SRL. Students, in turn, lack the self-assessment ability to capitalise on these opportunities. 

Only when we are able to weave together these different elements, will we be able to develop school-leavers who are ready to take on the challenges of the 21st century.

References

Australian Council of Deans of Education. (2001). New learning: a charter for Australian education (executive summary). Retrieved 5 February 2006 from: www.acde.edu.au
Boekaerts, M (1997). 'Self-regulated learning: a new concept embraced by researchers, policy makers, educators, teachers and students'. In Learning and instruction. 7(2), 161-186.
Boekaerts, M (2002). 'Bringing about change in the classroom: strengths and weaknesses of the self-regulated learning approach - EARLI Presidential Address, 2001'. In Learning and instruction. 12(6), 589–604.
Borkowski, JG, Weaver, CM, Smith, LE & Akai, CE (2004). 'Metacognitive theory and classroom practices'. In J. Ee, A. Chang & O. Tan (Eds.). Thinking about thinking: what educators need to know. Singapore: McGraw Hill.
Boud, D (2000). 'Sustainable assessment: rethinking assessment for the learning society'. In Studies in continuing education. 22(2), 151-167.
Flavell, JH (1987). 'Speculations about the nature and development of metacognition'. In FE Weinert & RH Kluwe (eds.). Metacognition, motivation and understanding. pp. 21-29). Hillside, New Jersey: Lawrence Erlbaum Associates.
Gonzales, P, Guzman, JC, Partelow, L, Pahlke, E, Jocelyn, L, Kastberg, D et al. (2004). 'Highlights from the Trends in International Mathematics and Science Study (TIMSS) 2003'. (electronic version). Retrieved 25 February 2006 from: nces.ed.gov.
Hattie, J, Biggs, J, & Purdie, N. (1996). 'Effects of learning skills interventions on student learning: A meta-analysis'. In Review of educational research. 66(2), 99-136.
Hoban, GF (2000). Teacher learning for educational change: a systems thinking approach. Buckingham, UK: Open University Press.
'Learning, Creating, Communicating: A Curriculum Review' (1998) (electronic version). Retrieved 13 March 2006 from: www.moe.gov.sg.
Northwest Regional Educational Laboratory. (2004). Developing Self-Directed Learners.
Palmer, SB & Wehmeyer, ML (2003). 'Promoting self-determination in early elementary school: teaching self-regulated problem-solving and goal-setting skills'. In Remedial and special education. 24(2), 115-126.
Pintrich, PR & DeGroot, EV (1990). 'Motivational and self-regulated learning components of classroom academic performance'. In Journal of educational psychology. 82(1), pp.33-40.
Schunk, DH (2001). 'Social cognitive theory and self-regulated learning'. In BJ Zimmerman & DH Schunk (eds.), Self-regulated learning and academic achievement: theoretical perspectives. 2nd ed., pp. 125-151. Mahwah, NJ. USA: Lawrence Erlbaum Associates Incorporated.
Schunk, DH & Zimmerman, BJ (1994). Self-regulation of learning and performance. Hillsdale: Lawrence Erlbaum.
Stiggins, RJ (1999). 'Assessment, Student Confidence, and School Success' (electronic version). Retrieved 15 February 2006 from: www.pdkintl.org.
Swanson, HL (2001). 'Research on interventions for adolescents with learning disabilities: A meta-analysis of outcomes related to higher-order processing'. In The elementary school journal. 101(3), 331-348.
Tan, K.(2004). 'Does student self-assessment empower or discipline students?'. In Assessment and evaluation in higher education. 29(6), 651-662.
Teach Less, Learn More (electronic (2007) version). Retrieved 6 April 2007 from: www.moe.gov.sg.
Towndrow, PA (2004). Teachers as digital task designers: an agenda for research and professional development. National Institute of Education.
Wiggins, G. (1993). 'Assessment: authenticity, context and validity'. In Phi delta kappan, 75(3), 210-214.
Zimmerman, BJ (2001). 'Theories of self-regulated learning and academic achievement: an overview and analysis'. In B. J Zimmerman & DH Schunk (eds.). Self-regulated learning and academic achievement: theoretical perspectives. (2nd ed., pp. 1-37). Mahwah, NJ, USA: Lawrence Erlbaum Associates, Incorporated.
Zimmerman, BJ (2002). 'Becoming a self-regulated learner: an overview'. In Theory into practice. 41(2), 64-70.
Zimmerman, BJ & Martinex-Pons, M (1986). 'Development of a structured interview for assessing student use of self-regulated learning strategies'. In American educational research journal. 23(4), 614-628.