High Educ (2015) 70:469–486

DOI 10.1007/s10734-014-9849-3

Promoting university students’ metacognitive regulation

through peer learning: the potential of reciprocal peer
tutoring

Liesje De Backer • Hilde Van Keer • Martin Valcke

Published online: 23 December 2014

 Springer Science+Business Media Dordrecht 2014

Abstract Although successful learning in university education can be advanced by

students’ competence to self-regulate their learning, students often possess insufficient
metacognitive regulation skills to regulate their learning adequately. The present study
investigates changes in university students’ adoption of metacognitive regulation after
participating in reciprocal peer tutoring (RPT). A quasi-experimental pretest–posttest
design was adopted, involving an experimental (n = 51) and two control groups; CG1
(n = 24) and CG2 (n = 22). Experimental students participated in a RPT intervention
during a complete semester. Metacognitive regulation was assessed by means of think-
aloud protocol analysis. Results indicate that RPT is promising to promote metacognitive
regulation. Experimental students increasingly adopt monitoring, evaluation, and orienta-
tion and significantly evolve towards deep-level regulation from pretest to posttest. Except
for an increased use of low-level comprehension monitoring, none of the evolutions in
experimental students’ regulation could be discerned for students in both control groups.

Keywords Metacognitive regulation
Peer tutoring
Think-aloud protocol analysis

Collaborative learning

Introduction

Metacognitive regulation is central to self-regulated learning and contributes to an

important extent to students’ performance (Meijer et al. 2006; Zimmerman and Schunk
2011). It generally advances academic success, especially in higher education since both
organisational structures and academic assignments at this educational level emphasise
self-management and independent learning (Nota et al. 2004). Nonetheless, students’
metacognitive regulation is often insufficient to self-regulate their learning adequately,

L. De Backer (&)
H. Van Keer
M. Valcke

Department of Educational Studies, Ghent University, H. Dunantlaan 2, 9000 Ghent, Belgium
e-mail: [Link]@[Link]

123
470 High Educ (2015) 70:469–486

revealing the necessity to design, implement, and evaluate initiatives fostering metacog-
nitive regulation.
The present study examines the impact of reciprocal peer tutoring (RPT) on university
students’ metacognitive regulation. Unlike most other research, it does not focus exclu-
sively on a particular regulation skill (e.g. monitoring), but takes an integrative perspective
when assessing students’ metacognitive regulation. It more specifically investigates
changes in students’ metacognitive orientation, planning, monitoring, and evaluation after
participation in RPT. By enhancing our understanding of the differential impact of RPT on
students’ adoption of regulation skills, the present study offers direct cues to optimally
foster students’ regulation through RPT.

Theoretical framework

Metacognitive regulation

Metacognition refers to both the awareness and active control students have over their
cognitive activities when engaged in learning or academic problem-solving (Brown 1987;
Pintrich 2004). The first component, metacognitive knowledge, concerns insights of stu-
dents about themselves as learners, insight into learning strategies, and insight into the
usefulness of these strategies within specific learning conditions (Pintrich 2004; Zim-
merman and Schunk 2011). The second component, metacognitive regulation, refers to
self-regulatory skills and strategies adopted by students to actively control and coordinate
their learning and performance (Greene and Azevedo 2009; Meijer et al. 2006). The focus
of the present study is on this second component of metacognition. We distinguish ori-
enting, planning, monitoring, and evaluating as key regulation skills (Brown 1987;
Veenman et al. 1997), which unfold over weakly sequenced and recursive problem-solving
phases (Greene and Azevedo 2009; Winne and Hadwin 1998). When orienting, students
engage in task analysis to get acquainted with learning objectives or task demands (Pintrich
2004). This can result in awareness of task perceptions or prior knowledge activation
(Meijer et al. 2006; Winne and Hadwin 1998). Planning encompasses selecting and
sequencing problem-solving strategies, allocating resources, and developing action plans
(Greene and Azevedo 2009; Meijer et al. 2006); either at the onset or during task execution
(e.g. after a subtask). Monitoring involves the online quality control of problem-solving,
aimed at identifying inconsistencies and optimising task execution (Meijer et al. 2006;
Moos and Azevedo 2009). Comprehension monitoring refers to control activities focusing
on the correctness of one’s understanding (Pintrich 2004; Veenman et al. 1997), whereas
monitoring of progress focuses on the adequateness of problem-solving strategies or the
quality of students’ progress (Greene and Azevedo 2009; Moos and Azevedo 2009).
Finally, evaluation involves learners’ self-judgement upon completion of problem-solving
(Veenman et al. 2005). This can be directed at both learning outcomes and the learning
process (Meijer et al. 2006).

Low-level versus deep-level metacognitive regulation

The present study distinguishes low-level and deep-level metacognitive regulation, intro-
ducing a more in-depth operationalisation. Low-level orientation is solely directed at
exploring task demands, whereas deep-level orientation aims at processing task demands
and activating prior knowledge (Veenman et al. 2005). Low-level planning implies the

123
High Educ (2015) 70:469–486 471

development of a single problem-solving plan, whereas deep-level planning involves

selecting a plan after considering various problem-solving alternatives (Meijer et al. 2006;
Veenman et al. 1997). When students check their progress or the comprehensiveness of
their understanding, they engage in low-level monitoring. Reflective comments on the
quality of their perceived progress or elaborative, thought-provoking reflections imply
deep-level monitoring (Moos and Azevedo 2009). Correspondingly, low-level evaluation
involves checking and commenting on either learning outcomes or process factors,
whereas deep-level evaluation implies reflective judgements on both (Veenman et al.
2005).

Optimising metacognitive regulation through peer tutoring

Fostering metacognitive regulation requires direct observation of explicitly modelled

metacognitive behaviour, along with explanations about regulation skills (Schunk and
Zimmerman 2007; Volet et al. 2009). Students should further be challenged to internalise
the modelled behaviour at the individual level, by practising and subsequently reflecting
upon a variety of self-regulatory strategies. Additionally, learners should be encouraged to
discuss and control their learning, refining their metacognitive regulation (Schunk and
Zimmerman 2007). Current research centres on modelling by peers during collaborative
learning (Volet et al. 2009). During collaborative learning, students discuss meaning and
compare their thinking with peers, requiring both regulation of their own cognition and
control of how peers learn (Volet et al. 2009). Despite growing consensus on the meta-
cognitive learning opportunities during collaborative learning, empirical evidence on its
differential influence on students’ adoption of specific regulation skills is limited. The
present study aims to enhance our understanding of the metacognitive potential of col-
laborative learning, by studying the specific impact of reciprocal peer tutoring on uni-
versity students’ use of orientation, planning, monitoring, and evaluation.
Peer tutoring is characterised by active academic helping and supporting between peers
in small groups or student pairs (Falchikov 2001; Topping 1996). One peer, the tutor, is
expected to take a direct pedagogical responsibility by creating learning opportunities
through questioning, clarifying, and scaffolding (Chi et al. 2001). The students being
cognitively challenged by this peer tutor are called tutees. Reciprocal peer tutoring (RPT)
is characterised by the structured exchange of the tutor role among peers in the PT groups/
pairs (Topping 1996). Although the available research illustrates that peer tutoring chal-
lenges students’ metacognitive regulation and particularly promotes their adoption of
monitoring (e.g. King 1997; Roscoe and Chi 2008), possible effects on other regulation
skills are underexposed.

Aim of the study and research hypotheses

The present study aims to study the changes in university students’ adoption of meta-
cognitive regulation skills after participation in RPT. Since literature provides evidence for
metacognitive learning opportunities within collaborative learning, we hypothesise that
RPT will have a positive impact on students’ use of metacognitive regulation skills
(hypothesis 1). During tutoring, conceptual discussions can take place at different levels of
social interaction: the individual, dyadic, and group level (Volet et al. 2009). Conse-
quently, students experience the need to metacognitively regulate their own, each other’s,
or the group’s cognition, increasing chances for metacognitive engagement. Because more

123
472 High Educ (2015) 70:469–486

regulatory control often results in the use of profound metacognitive strategies (Chin and
Brown 2000; Greene and Azevedo 2009), we additionally hypothesise that RPT will
promote students’ involvement in deep-level metacognitive regulation (hypothesis 2).

Method

Design

A quasi-experimental pretest–posttest design was adopted, involving an experimental and

two control groups. Experimental students participated in the RPT intervention during a
complete semester. Both at the start (October) and the end (December) of the semester, the
regulation skills of individual participants (n = 97) were assessed.

Participants and setting

The study was conducted in a naturalistic university setting. The experimental group (EG)
consisted of the complete population of 64 first-year students, Educational Sciences, who
already obtained a professional bachelor degree (12.5 % males and 87.5 % females).1 This
gender distribution is representative for the Educational Sciences students population in
Flanders (Belgium). Unlike other studies (e.g. Cheng and Ku 2009; Duran and Monereo
2005), the present study did not aim to compare RPT with a different type of student-
activating learning (e.g. fixed peer tutoring, self-explaining). Therefore, we preferred not to
assign students of the Educational Sciences programme with a professional bachelor
degree to either an experimental or control treatment, due to ethical reasons. Such a design
would imply students being deprived from the benefits of peer-assisted learning, inherent
to RPT. Alternatively, we involved two control groups. The first (CG1) consisted of 24
freshmen in the Educational Sciences programme (12.5 % males and 87.5 % females).
Despite differences in participants’ age and prior experience in higher education, these
students are enrolled in the same curriculum as EG students. This curriculum aims at
introducing students into the basic theoretical frameworks of Educational Sciences and
consists of general social sciences courses, specialised pedagogical courses, and method-
ology courses. The second control group (CG2) consisted of 22 first-year students in the
Social Welfare Studies programme of the same faculty, who also attained a professional
bachelor degree (14.3 % males and 85.7 % females). Although these students are enrolled
in a somewhat different curriculum, their background and prior experience in higher
education is comparable to EG students. CG2 students’ curriculum aims at introducing
students into the theoretical frameworks of Social Welfare Studies and consists of general
pedagogical courses, specialised courses on Social Welfare Studies, and methodology
courses. The didactical approach of these courses is comparable to the curriculum of EG
and CG1 students: a dominance of theoretical lectures is alternated with group work, which
focuses on writing a theoretical paper with a small group of students. During the research
period, students from both control groups attended regular curriculum activities as part of
their academic training. They were not involved in tutoring, neither in any comparable

1
Although all 64 RPT students participated in the pretest, not all of them attended the posttest assessment.
Additionally, the tape recordings of some RPT students demonstrated technical problems. Therefore, the
data of 13 EG students were excluded from the analysis.

123
High Educ (2015) 70:469–486 473

collaborative learning approach, encompassing systematic conceptual interaction and

discussions among students in a face-to-face setting.
EG students were randomly assigned to eleven RPT groups. The RPT programme was a
formal component of the 5-credit course ‘‘Instructional Sciences’’ (of which the theoretical
lectures were part of both EG and CG1 students’ curriculum). Informed consent was
obtained from all participants in the study.

Intervention

The RPT intervention consisted of eight successive face-to-face sessions (each taking 2 h),
in which students tutored each other in small and stable groups of six. The tutor role was
changed at each session. As a manipulation check, all RPT groups were observed weekly,
to control whether students enacted their roles adequately.

RPT assignments

During each session, students worked on authentic, open-ended group assignments (De
Backer et al. 2012). Each assignment comprised of an outline of learning objectives to
guide peers’ discussions to central course-related topics; a subtask aimed at getting familiar
with the theme-specific terminology; and a subtask in which students applied theoretical
notions to realistic instructional cases.

Training

All EG students participated in a compulsory tutor training 1 week before the onset of the
RPT intervention, focussing on tutor responsibilities and generic tutoring skills (i.e.
establishing a safe learning climate, managing peer interactions, questioning, explaining,
scaffolding, and providing feedback) (Chi et al. 2001; King 1997). The training used
videotaped examples of good and bad practices, which were discussed in depth, role plays
in which students experienced multiple tutor responsibilities and received feedback on
their tutoring approach, and the in-depth analysis of authentic case studies focusing on
specific tutor competences. The outlines of the training were presented in a manual pro-
vided to all EG students.

Tutor guide

To prepare themselves for the tutor role, tutoring students received a session-specific ‘‘tutor
guide’’ 1 week in advance. This offered additional information about the theoretical
learning content to focus upon in the RPT session and assured a difference in tutors’ and
tutees’ domain-specific knowledge (Topping 1996). Although the theoretical content of the
guide differed across sessions, its structure and design were identical throughout the RPT
intervention. In addition to offering theoretical knowledge, the guide exemplified a step-
wise problem-solving approach (De Backer et al. 2012).

Interim support

An interim supervision session and two-weekly feedback sessions were organised, to

provide tutor support (Falchikov 2001). The supervision was set up in small groups of

123
474 High Educ (2015) 70:469–486

twelve students and directed by a university staff member, who encouraged students to
reflect on the adequacy of their tutor/tutee behaviour. Additionally, the staff member
provided group-specific feedback once every 2 weeks.

Instruments

Assessment of students’ metacognitive regulation was based on think-aloud protocol

analysis. At the start and the end of the research, all participants individually performed an
academic task that was videotaped. Participants were instructed to solve the task and to
verbalise their problem-solving, resulting in verbal protocols (Greene et al. 2011).

Task

The individual think-aloud task comprised of a theoretical text and a real-life case relevant
to the course ‘‘Instructional Sciences’’ (i.e. topic ‘‘evaluation’’ and ‘‘social inequity in
education’’ at pretest and posttest, respectively). Students solved thought-provoking
questions about the text. In case students stopped verbalising during task performance, they
were prompted by the assessor to continue thinking aloud. Apart from a difference in the
central topic, all aspects of the think-aloud task and the assessment procedure were
identical at pretest and posttest. Both tasks provided a challenging yet comprehensible text,
comprised of multiple parts, and had to be executed during the mild time constraint of
30 min (Greene et al. 2011).

Coding instrument

To analyse students’ verbal protocols, we developed a literature-based coding instrument,

representing a multi-layered model of metacognitive regulation (De Backer et al. 2012).
Orientation, planning, monitoring, and evaluation are adopted as the main coding cate-
gories and further specified with sub-coding categories (i.e. task analysis, content orien-
tation, planning in advance, interim planning, comprehension monitoring, monitoring of
progress, evaluation of learning outcomes, and evaluation of the learning process).
Additionally, a dimension on the low-/deep-level approach is included.

Coding strategy

The verbal protocols of all students were transcribed verbatim and coded by two inde-
pendent and trained coders. They double coded 25 % of the protocols to determine
interrater reliability. Cohen’s kappa (j = 0.78) indicates high overall interrater reliability
and good agreement beyond chance for the four main coding categories (jorientation = 0.89,
jplanning = 0.86, jmonitoring = 0.82, and jevaluation = 0.89). The coding procedure followed
subsequent phases and focused exclusively on students’ verbalised behaviour. Each verbal
protocol was initially segmented according to changes in the verbalisation focus, which
formed the boundaries of substantial ‘‘episodes’’ (Chi et al. 2001). An episode is con-
ceptualised as a brief segment of the overall verbal protocol that was centred around one
particular action. After segmentation, each episode received a general code, indicating
whether it concerned a metacognitive, task-executive, or off-task episode. Second, meta-
cognitive episodes were selected for more detailed ‘‘statement coding’’ (Roscoe and Chi
2008). A statement refers to a single thematically consistent verbalisation of a single

123
High Educ (2015) 70:469–486 475

metacognitive action. Next, the identified metacognitive statements were coded by means
of the developed coding instrument. Every statement received a general code (indicating
the regulation skill it addressed) and a more differentiated code (referring to the concre-
tised regulation strategies situated in the subcategories of the coding instrument). Addi-
tionally, the approach (low level /deep level) to the identified metacognitive statements
was coded. Table 1 exemplifies the coding strategy.

Data analysis

First, the frequency of occurrence of all regulation skills and strategies, and of low-/deep-
level approaches, was calculated for each protocol per participant. The frequency of
occurrence of regulation skills and approaches in individual students’ protocols was
aggregated for each research condition. These frequencies per condition were used for
analysis purposes. Second, pre-analysis investigations were conducted to check both the
assumption of normally distributed data (i.e. by means of Kolmogorov–Smirnov tests) and
the assumption of homogeneity of variance (i.e. by means of Levene’s tests). Since the
results revealed that neither assumption was violated, parametric analyses (i.e. t tests) were
performed to examine the significance of differences in the adoption of metacognitive
regulation between conditions and measurement occasions. Third, independent samples
t tests were conducted (i.e. comparing EG with CG1, EG with CG2, and CG1 with CG2,
respectively) to assure that the metacognitive regulation of students in the three conditions
was comparable at pretest. Fourth, the frequency of occurrence of regulation skills and
approaches at pretest and posttest was compared by means of paired samples t tests, to
study the impact of RPT on students’ use of metacognitive skills (hypothesis 1) and on
their deep-level regulation approach (hypothesis 2). Significant differences in students’
metacognitive regulation at pretest and posttest were tested for each condition separately
(see Tables 3, 4, 5). Cohen’s d is reported as a measure of effect size of significant
differences which were identified in students’ adoption of regulation skills and a deep-level
regulation approach. The significance level was 0.05 for all analyses.

Results

Descriptive analysis on students’ metacognitive regulation

Table 2 demonstrates a dominant involvement of all students in metacognitive monitoring,

both at pretest and posttest, whereas their adoption of orientation, planning, and evaluation is
rather limited, especially at pretest. Results moreover reveal a clear pretest-to-posttest change
towards more frequent use for the majority of metacognitive strategies by EG students. For
some metacognitive behaviour, this trend is markedly smaller in both control groups (e.g.
comprehension monitoring, evaluation of learning outcomes), while for other metacognitive
strategies this positive evolution cannot be discerned at all (e.g. orientation, monitoring of
progress). In contrast, the planning behaviour of students appears to evolve similarly for all
research groups. Results further reveal students’ limited attention to evaluation of the
learning process, deep-level monitoring of progress, deep-level planning, and taking into
account task perceptions. This is observed at pretest and posttest, in all conditions.2

2
Metacognitive regulation strategies with very low frequency of occurrence (\1 % in Table 2) were
removed from further analyses.

123
476 High Educ (2015) 70:469–486

Table 1 Exemplified coding strategy

Verbal protocol excerpt Episode Statement coding Approach
coding

It is about assessment and evaluation Metacognition Exploring task demands Low

(orientation) level
I first read the instructions to know what I have Metacognition Exploring task demands Low
to do. [student reads silently] (orientation) level
Okay, search for different evaluation forms. Metacognition Processing task demands Deep
And the second question, tell for which (orientation) level
purposes assessment is used
Now I will read the text Metacognition Formulating problem- Low
solving plan (planning) level
[student reads] ‘‘Regarding the functions of Task
evaluation, we distinguish summative and execution
formative evaluation. Summative occurs at the
end of learning, aimed at controlling whether
objectives are reached. Formative evaluation
intends to provide interim feedback…’’
[student does not finish sentence]
Wait (…) there is summative and formative… Metacognition Noting lack of Low
but what is the difference? comprehension level
(Monitoring)
I reread this part. [student rereads silently] Task
execution
So summative at the end, formative during Metacognition Demonstrating Low
learning comprehension by level
repeating (Monitoring)
I continue reading (…) Task
execution

Impact of RPT on students’ adoption of metacognitive regulation

Paired samples t tests on EG students’ adoption of regulation skills reveal a significant

increase in monitoring, evaluation, and (albeit to a lesser extent) orientation from pretest to
posttest (see Table 3). Additionally, EG students demonstrate a significantly enhanced
involvement in deep-level regulation at posttest. The reported effect sizes for these evo-
lutions are moreover large (see Table 3). No significant changes could be identified in none
of the above-mentioned regulation skills, neither in a deep-level regulation approach, for
CG1 students (see Table 4) or CG2 students (see Table 5). The planning behaviour of
students in all conditions does not differ significantly from pretest to posttest.
The following paragraph outlines the pretest-to-posttest changes in both the frequency
of occurrence of metacognitive regulation skills (hypothesis 1) and adoption of deep-level
regulation (hypothesis 2) in more detail. Significant changes are addressed for each
metacognitive regulation skill separately.

Monitoring

Table 3 reveals a major increase in EG students’ adoption of comprehension monitoring

from pretest to posttest. Compared to the pretest, EG students demonstrate enhanced
involvement in monitoring their understanding at posttest, both by paraphrasing

123
 Table 2 Occurrence of metacognitive regulation at pretest and posttest for the three research conditions (frequenciesa and percentagesb)
Metacognitive regulation Pretest Posttest

EG (n = 51) CG1 (n = 24) CG2 (n = 22) EG (n = 51) CG1 (n = 24) CG2 (n = 22)

Freq. % Freq. % Freq. % Freq. % Freq. % Freq. %

Orientation 70 10.9 28 10.0 37 14.5 220 10.3 38 8.4 35 8.3

Task analysis 64 9.9 26 9.3 34 13.3 117 8.3 38 8.4 32 7.6
High Educ (2015) 70:469–486

Exploring task demands (LL) 54 8.4 22 7.9 29 11.4 130 6.1 35 7.7 29 6.9
Processing task demands (DL) 10 1.5 4 1.4 5 1.9 47 2.2 3 0.7 3 0.7
Content orientation 5 0.8 2 0.7 3 1.2 35 1.6 0 0.0 1 0.2
Generating hypotheses (DL) 1 0.2 0 0.0 1 0.4 2 0.1 0 0.0 0 0.0
Activating prior knowledge (DL) 4 0.6 2 0.7 2 0.8 33 1.5 0 0.0 1 0.2
Becoming aware of task perceptions 0 0.0 0 0.0 0 0.0 8 0.3 0 0.0 2 0.5
Planning 88 13.7 52 18.6 42 16.5 211 9.9 106 23.4 100 23.7
Planning in advance 28 4.3 14 5.0 17 6.7 76 3.6 29 6.4 27 6.4
Formulating problem-solving plan (LL) 28 4.3 14 5.0 17 6.7 76 3.6 29 6.4 27 6.4
Selecting problem-solving plan (DL) 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0
Interim planning 60 9.5 38 13.6 25 9.8 135 6.3 78 17.0 73 17.3
Formulating problem-solving plan (LL) 60 9.5 38 13.6 25 9.8 125 6.3 78 17.0 73 17.3
Selecting problem-solving plan (DL) 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0
Monitoring 447 69.4 191 68.2 162 63.5 1,472 68.8 286 63.1 263 62.3
Comprehension monitoring 228 35.4 84 30.0 70 27.5 1,019 47.7 189 41.7 202 47.9
Noting lack of comprehension 85 13.2 42 15.0 36 14.4 138 6.5 38 8.4 33 7.8
Summarising main ideas 19 3.0 0 0.0 9 3.5 38 1.8 18 4.0 39 9.2
Demonstrating comprehension by repeating (LL) 95 14.8 28 10.0 19 7.5 445 20.8 105 23.3 89 21.1
Demonstrating comprehension by elaborating (DL) 33 5.1 14 5.0 6 2.4 395 18.5 29 6.4 41 9.8
Monitoring of progress 219 34.0 109 38.9 91 35.7 457 21.4 97 21.4 61 14.4
Checking of progress (LL) 217 33.7 108 38.6 89 34.9 402 18.8 95 21.0 59 14.0
477

123
 Table 2 continued
478

Metacognitive regulation Pretest Posttest

EG (n = 51) CG1 (n = 24) CG2 (n = 22) EG (n = 51) CG1 (n = 24) CG2 (n = 22)

123
Freq. % Freq. % Freq. % Freq. % Freq. % Freq. %

Reflecting on progress (DL) 2 0.3 1 0.3 2 0.8 55 2.6 2 0.4 2 0.4

Evaluation 33 5.1 3 3.2 13 5.1 231 10.7 21 4.6 23 5.5
Evaluating learning outcomes 33 4.9 9 3.2 13 5.1 220 10.3 21 4.6 23 5.5
Checking learning outcomes (LL) 32 0.2 9 3.2 13 5.1 206 9.6 20 4.4 21 5.0
Elaborating on learning outcomes (DL) 1 0.0 0 0.0 0 0.0 14 0.7 1 0.2 2 0.5
Evaluating learning process 0 0.0 0 0.0 0 0.0 11 0.4 0 0.0 0 0.0
Commenting on learning process (LL) 0 0.0 0 0.0 0 0.0 11 0.4 0 0.0 0 0.0
Reflecting on learning process (DL) 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0
LL low level, DL deep level
a
The columns ‘‘frequency’’ present the absolute frequency of occurrence of a particular metacognitive regulation skill (i.e. orientation, planning, monitoring, or evaluation) or
strategy (i.e. at the more concretised level of the coding instrument) within a research condition (i.e. EG, CG1, or CG2) at a particular measurement occasion (i.e. pretest or
posttest)
b
The columns ‘‘percentages’’ present the proportion of a particular metacognitive regulation skill or strategy within a research condition at a particular measurement occasion
taking into account all metacognitive regulation utterances that occurred and were assessed in that respective research group at that respective measurement occasion
High Educ (2015) 70:469–486
High Educ (2015) 70:469–486 479

Table 3 Results of pre- and posttest think-aloud protocol analysis in EG condition: Occurrence of meta-
cognitive regulation skills and low-level versus deep-level regulation
Metacognitive skill Frequency t(df) p Cohen’s
d
Pretest Posttest

M3 SD M SD

Orientation 1.37 1.21 4.21 2.06 -9.10 (50) \.001 1.73

Task analysis 1.25 1.09 3.33 1.69 -8.93 (50) \.001 1.53
Exploring task demands (LL) 1.06 0.18 2.49 1.02 -8.37 (50) \.001 1.55
Processing task demands (DL) 0.18 0.07 0.86 0.11 -5.69 (50) \.001 1.02
Content orientation 0.10 0.05 0.67 0.08 -5.80 (50) \.001 1.19
Activating prior knowledge (DL) 0.08 0.04 0.62 0.08 -5.58 (50) \.001 1.17
Planning 1.72 1.08 2.16 1.06 -1.91 (50) .062 0.40
Monitoring 8.37 3.33 28.86 8.66 -15.55 (50) \.001 3.14
Comprehension monitoring 4.10 2.39 19.98 7.14 -14.35 (50) \.001 3.33
Noting lack of comprehension 1.67 0.99 2.71 1.73 -3.85 (50) \.001 0.76
Summarising main ideas 0.37 0.10 0.74 0.11 -2.72 (50) \.001 0.47
Demonstrating comprehension by 1.47 0.17 8.73 2.06 -16.73 (50) \.001 3.38
repeating (LL)
Demonstrating comprehension by 0.65 0.14 7.74 4.08 -11.94 (50) \.001 2.79
elaborating (DL)
Monitoring of progress 4.29 1.75 8.96 3.43 -9.26 (50) \.001 1.79
Checking progress (LL) 4.29 1.79 7.88 3.02 -8.25 (50) \.001 1.49
Reflecting on progress (DL) 0.04 0.03 1.06 0.90 -7.38 (50) \.001 1.72
Evaluation 0.65 0.14 3.53 2.09 -13.42 (50) \.001 2.58
Evaluating learning outcomes 0.65 0.13 3.31 1.19 -14.06 (50) \.001 4.03
Checking outcomes (LL) 0.65 0.14 3.04 1.70 -13.98 (50) \.001 2.59
Elaborating on outcomes (DL) 0.00 0.00 0.27 0.07 -3.68 (50) .010 1.03
M refers to how often an individual student on average uses a metacognitive regulation skill during think-
aloud problem-solving at pretest and at posttest

information and by elaborating on the learning content. Although CG1 students and CG2
students also increase significantly in more comprehension monitoring through para-
phrasing at posttest (see Tables 4, 5, respectively), the pretest-to-posttest change is sig-
nificantly larger for EG students. Regarding students’ deep-level regulation approach, the
results reveal a major pretest-to-posttest change for deep-level comprehension monitoring
for EG students (see Table 3). EG students are, moreover, nearly equally involved in low-
level and deep-level comprehension monitoring, whereas control students predominantly
engage in low-level comprehension monitoring (see Table 2). CG1 students’ and CG2
students’ adoption of deep-level comprehension monitoring does not change significantly
from pretest to posttest (see Tables 4, 5, respectively).
Further, EG students significantly increase the frequency of monitoring their progress
from pretest to posttest (see Table 3). Remarkably, a significant negative pretest-to-posttest
change was revealed for both CG1 students (see Table 4) and CG2 students (see Table 5).
Although Table 3 additionally demonstrates a significant pretest-to-posttest increase in
deep-level monitoring of progress for EG students, their involvement in this respect

123
480 High Educ (2015) 70:469–486

Table 4 Results of pre- and posttest think-aloud protocol analysis in CG1 condition: Occurrence of
metacognitive regulation skills and low-level versus deep-level regulation
Metacognitive skill Frequency t(df) p Cohen’s
d
Pretest Posttest

M SD M SD

Orientation 1.17 1.09 1.58 1.17 -1.22 (23) 0.233 0.36

Task analysis 1.08 1.00 1.58 1.17 -1.52 (23) 0.143 0.45
Exploring task demands (LL) 0.92 0.58 1.46 0.83 -2.60 (23) 0.061 0.76
Processing task demands (DL) 0.17 0.13 0.12 0.09 0.25(23) 0.802 0.40
Content orientation 0.08 0.05 0.00 0.00 1.45 (23) 0.162 0.59
Activating prior knowledge (DL) 0.08 0.05 0.00 0.00 1.45 (23) 0.162 0.59
Planning 2.17 1.34 2.50 0.88 -1.09 (23) 0.286 0.30
Monitoring 7.96 2.71 7.50 1.69 1.06 (23) 0.302 0.21
Comprehension monitoring 3.50 1.79 5.12 1.72 -3.43 (23) 0.002 0.87
Noting lack of comprehension 1.75 1.11 1.08 0.87 2.56 (23) 0.071 0.43
Summarising main ideas 0.00 0.00 0.67 0.15 -4.29 (23) \.001 1.75
Demonstrating comprehension by 1.17 1.00 2.71 1.62 -4.21 (23) \.001 1.17
repeating (LL)
Demonstrating comprehension by 0.50 0.22 0.67 0.16 -0.59 (23) 0.558 0.18
elaborating (DL)
Monitoring of progress 4.54 1.69 2.33 1.40 6.20 (23) \.001 1.45
Checking progress (LL) 4.50 1.69 2.33 1.40 6.11 (23) \.001 1.39
Reflecting on progress (DL) 0.04 0.03 0.00 0.00 1.00 (23) 0.328 0.40
Evaluation 0.37 0.13 0.58 0.13 -1.00 (23) 0.328 0.32
Evaluating learning outcomes 0.37 0.13 0.58 0.13 1.00 (23) 0.328 0.32
Checking outcomes (LL) 0.37 0.13 0.58 0.13 -1.00 (23) 0.328 0.16
Elaborating on outcomes (DL) 0.00 0.00 0.00 0.00 – – –

remains low (2.6 %). CG1 students’ and CG2 students’ deep-level monitoring of progress
does not change significantly from pretest to posttest (see Tables 4, 5, respectively).

Evaluation

Table 3 shows a significant pretest-to-posttest increase in EG students’ evaluation of

learning outcomes. In contrast, CG1 students’ and CG2 students’ engagement in evaluating
learning outcomes remains limited and does not differ significantly from pretest to posttest
(see Tables 4, 5, respectively). Despite significant pretest-to-posttest differences in EG
students’ deep-level evaluation (see Table 3), the occurrence of the latter remains very
limited (0.7 %). No significant pretest-to-posttest changes could be found for CG1 stu-
dents’ and CG2 students’ deep-level evaluation (see Tables 4, 5, respectively).

Orientation

Although EG students significantly increase their task analysis and prior knowledge
activation from pretest to posttest (see Table 3), their involvement in orientation is less
pronounced compared to the changes in their monitoring (i.e. of comprehension and

123
High Educ (2015) 70:469–486 481

Table 5 Results of pre- and posttest think-aloud protocol analysis in CG2 condition: occurrence of
metacognitive regulation skills and low-level versus deep-level regulation
Metacognitive skill Frequency t(df) p Cohen’s
d
Pretest Posttest

M SD M SD

Orientation 1.68 1.39 1.82 1.53 -0.37 (21) 0.715 0.11

Task analysis 1.54 1.22 1.64 1.29 -0.28 (21) 0.785 0.07
Exploring task demands (LL) 1.32 0.83 1.45 1.10 -0.48 (21) 0.633 0.14
Processing task demands (DL) 0.23 0.11 0.18 0.08 0.44 (21) 0.665 0.10
Content orientation 0.14 0.09 0.09 0.06 0.44 (21) 0.665 0.12
Activating prior knowledge (DL) 0.09 0.05 0.09 0.06 0.00 (21) 0.999 0.00
Planning 1.91 1.44 1.86 0.99 0.13 (21) 0.894 0.04
Monitoring 7.36 2.82 9.04 1.56 -2.18 (21) 0.061 0.76
Comprehension monitoring 3.18 2.03 6.91 1.79 -5.93 (21) \.001 1.94
Noting lack of comprehension 1.64 0.72 1.00 0.81 2.85 (21) 0.010 0.82
Summarising main ideas 0.41 0.13 1.36 1.00 -3.47 (21) 0.002 1.19
Demonstrating comprehension by 0.86 0.27 3.36 1.61 -5.56 (21) \.001 1.72
repeating (LL)
Demonstrating comprehension by 0.32 0.14 1.18 0.38 -2.09 (21) 0.069 0.49
elaborating (DL)
Monitoring of progress 4.14 2.31 2.09 1.26 3.70 (21) \.001 1.14
Checking progress (LL) 4.14 2.31 2.04 1.17 3.88 (21) \.001 1..19
Reflecting on progress (DL) 0.09 0.06 0.00 0.00 1.45 (21) 0.162 0.61
Evaluation 0.59 0.21 0.50 0.12 0.36 (21) 0.724 0.11
Evaluating learning outcomes 0.64 0.22 0.50 0.13 0.51 (21) 0.505 0.22
Checking outcomes (LL) 0.64 0.22 0.50 0.13 0.53 (21) 0.613 0.16
Elaborating on outcomes (DL) 0.14 0.09 0.00 0.00 1.37 (21) 0.186 0.58

progress) and evaluation (i.e. of learning outcomes). Table 3 further reveals a significant
increase in EG students’ adoption of both low-level and deep-level orientation. However,
EG students’ engagement in deep-level orientation remains limited (3.5 %). No significant
pretest-to-posttest changes could be found for CG1 students’ and CG2 students’ deep-level
orientation (see Tables 4, 5, respectively).

Discussion

Impact of RPT on students’ adoption of metacognitive regulation

The results indicated that students are predominantly involved in monitoring their prob-
lem-solving when executing an academic task, both at pretest and at posttest, in all
research conditions. The findings further revealed an increased adoption of monitoring and
to a lesser extent evaluation and orientation by EG students at posttest. Except for low-
level comprehension monitoring, the above-mentioned pretest-to-posttest increases were
not discerned in the control conditions. These results consequently demonstrated that RPT

123
482 High Educ (2015) 70:469–486

is more beneficiary for fostering students’ adoption of monitoring and (albeit to a lesser
extent) evaluation and orientation, compared to traditional teaching approaches. A sig-
nificant impact on students’ planning behaviour could, however, not be distinguished, not
for EG students neither for the control conditions. This might be due to the design of the
think-aloud task, for the latter partially determines the outcomes of protocol analysis
(Greene et al. 2011). Since students were expected to solve three thought-provoking
questions on a well-structured academic task, the opportunities to plan were probably
scarce. Additionally, students might not have felt the need to sequence problem-solving
steps within the available time framework. It should further be noted that the developed
coding instrument might not have been sensitive enough to capture students’ planning
appropriately. Future research should aim to conceptualise metacognitive planning in a
more specific way in order to optimise its assessment.
RPT appeared to have a critical impact on students’ application of comprehension
monitoring. During RPT, students were challenged to approach the learning content crit-
ically and to negotiate its meaning, resulting in self-questioning (Roscoe and Chi 2008). It
seems plausible to assume that a semester-long experience in this cognitively challenging
RPT environment prompted students to internalise this comprehension monitoring
behaviour. It should be noted, however, that control students also checked their under-
standing more often at posttest. Consequently, students’ increased involvement in com-
prehension monitoring might be partially attributed to their experienced need for self-
regulation in higher education (Nota et al. 2004). During their first semester at university,
all students were presumably faced with the demands for elaborative thinking and self-
control of one’s understanding, resulting in the adoption of monitoring strategies. Nev-
ertheless, our findings demonstrated that EG students showed a significant larger adoption
of monitoring strategies as compared to students in control conditions, implying an added
value of RPT.
Our findings further revealed increased monitoring of progress and more evaluative
comments on learning outcomes for EG students. In contrast, both control groups did
not increase their metacognitive reflection and evaluation, not during the course of
problem-solving (i.e. monitoring their progress), neither upon completing it (i.e.
metacognitive evaluation). These findings suggest that RPT is a promising approach
when aiming to advance students’ evaluative reflections. It can be assumed that the key
elements of PT (i.e. asking and answering thought-provoking questions, providing
knowledge-building explanations, scaffolding, giving feedback.) directly fostered stu-
dents’ self-reflections and evaluative insights (Chi et al. 2001; Roscoe and Chi 2008).
Additionally, the design of the RPT learning materials might have promoted students’
evaluative engagements, for the assignments and weekly tutor guides systematically
outlined learning objectives, which might have served as an evaluative tool (Zimmer-
man and Schunk 2011).
Given EG students’ increased adoption of certain metacognitive regulation strategies,
we recommend higher education instructors to implement RPT when aiming to promote
university students’ metacognitive regulation. The results more specifically revealed that
training students to tutor each other stimulates them to start monitoring their learning more
frequently. Additionally, the present study suggests that well-structured and goal-oriented
group assignments have potential to elicit particular evaluation strategies (albeit less
pronounced compared to evoking monitoring acts). Consequently, organising RPT requires
educators to carefully design learning materials which can encourage students in regulating
their learning.

123
High Educ (2015) 70:469–486 483

Impact of RPT on students’ deep-level regulation approach

Notwithstanding students’ dominant use of low-level regulation, at pretest and posttest, our
findings revealed significant effects of RPT on students’ deep-level metacognitive regu-
lation. Given that none of the control groups demonstrated significant pretest-to-posttest
changes towards a deep-level regulation approach for none of the key regulation skills, the
above-mentioned result implies that RPT is more beneficiary to enhance students’
engagement in deep-level regulation, compared to traditional teaching approaches. Our
findings more specifically demonstrated that EG students outperformed control students in
the adoption of deep-level comprehension monitoring. Since tutors were trained to promote
tutees’ profound reflective thinking by asking critical questions, providing cognitive
scaffolds, and giving knowledge-building explanations, it could be assumed that RPT
participants observed and eventually internalised these strategies, modelled by their tutors.
EG students additionally revealed an increased use of deep-level task analysis, profound
monitoring of progress, and deep-level evaluation of learning outcomes. It should be noted,
however, that the frequency of occurrence of these deep-level regulation strategies
remained low. Students’ rather limited involvement in deep-level regulation might be
explained by both the need for explicit metacognitive prompts and more extensive
opportunities to practise regulation skills (Schunk and Zimmerman 2007). Since orienta-
tion and evaluation can only be employed before and upon completion of task execution,
respectively, their frequency of occurrence might have been too limited for students to
evolve towards frequent deep-level orientation and evaluation.
Based on our results, we advise higher education instructors to implement long-lasting
RPT interventions, allowing students the time they need to evolve towards more frequent
practice with different deep-level regulation strategies. This practice could increase the
chances of students starting to internalise a deep-level approach when regulating their
learning. Additionally, it might be advisable to include explicit scaffolds in the RPT
learning materials, which can directly prompt students’ deep-level regulation during all
phases of problem-solving. Such scaffolds might particularly be relevant for evoking
orientation, evaluation, and planning, given that these regulation skills remained rather
low-level in the current RPT intervention, which was characterised by open-ended RPT
assignments, not directing students’ regulation. Prompting students’ deep-level regulation
approach in future research could enhance their regulative engagement in this respect,
allowing students to refine and spontaneously adopt a deep-level approach when applying
diverse regulation skills.

Limitations and recommendations for future research

Since the present study was conducted in a naturalistic university setting, it was, due to
ethical reasons, preferable not to randomly assign students from the same class group to
either the experimental or control condition. Although two control groups were involved,
these were not completely comparable due to differences in participants’ background or
their university curriculum. Consequently, caution is needed when interpreting the sig-
nificant changes in students’ metacognitive regulation. Alternative explanations for EG
students’ increased regulation can be found in both students’ cognitive gains and the
emphasis on self-regulation in the course ‘‘Instructional Sciences’’. Empirical research
demonstrated that students’ metacognitive regulation is correlated to their cognitive actions
and performance (Greene and Azevedo 2009; Zimmerman and Schunk 2011). Students
with higher levels of general and domain-specific knowledge are expected to demonstrate a

123
484 High Educ (2015) 70:469–486

larger involvement in regulation, often resulting in better performance. Moreover, high

levels of knowledge and academic experience are assumed to positively influence the
quality of learners’ metacognitive skills (Chin and Brown 2000). This study was set up in
relation to the course ‘‘Instructional Sciences’’ introducing students in theories about
learning and instruction, including self-regulation. The particular course context might
have enhanced EG students’ problem-solving awareness, resulting in a theory-driven
execution of the think-aloud task. It should be noted, however, that metacognitive gains
based on students’ knowledge of ‘‘Instructional Sciences’’ could equally be expected from
CG1 students, since the course was also a formal part of their curriculum. Nevertheless, EG
students’ increased regulation at posttest was not reflected in the regulation behaviour of
CG1 students.
Furthermore, the assessment of students’ metacognitive regulation was exclusively
based on their verbalised problem-solving. It can be assumed, however, that students do
not always explicitly articulate their thinking and regulation (Meijer et al. 2006), implying
that the identification of metacognitive utterances in the think-aloud protocols might not
have been exhaustive. Another limitation of think-aloud protocol analysis concerns the risk
of reactivity (Greene et al. 2011), since asking students to verbalise can increase their
attention to their cognitive processing, possibly resulting in more regulation than they
would spontaneously demonstrate. Data triangulation through multiple concurrent
assessment techniques might be more preferable in future research (Meijer et al. 2006).
Although the results suggested a positive impact of RPT, we only reported short-term
effects. Long-lasting interventions are needed to guarantee an enduring impact. Further-
more, the time-consuming nature of think-aloud protocol coding only allowed for data
analysis on a relatively small sample (Roscoe and Chi 2008). The study was moreover
conducted in a particular setting, with students studying a specific course. Future research
preferably involves other student populations, alternative instructional settings, or other
tutoring formats to increase the representativeness of the findings.
It should further be noted that although the RPT intervention successfully elicited
students’ metacognitive monitoring, its effects on the adoption of orientation and evalu-
ation, as well as of deep-level metacognitive regulation (i.e. monitoring and orientation)
were less prominent. This might be related to the format of the think-aloud tasks and RPT
assignments (Greene et al. 2011; Perry and Winne 2013). Both might have stimulated
students to particularly check their comprehension and progress but might have been less
appropriate to evoke (and assess) other regulative acts. Future research with alternative
task formats is needed to examine whether more changes can be discerned in students’
adoption of specific regulation skills and approaches after participation in RPT.
Given the observation that the impact of RPT might be stimulated by specific collab-
oration patterns within particular tutoring groups (Chi et al. 2001), future process-oriented
investigations could also be promising (Roscoe and Chi 2008). Analysis of videotaped
RPT sessions could moreover complement the present findings. Directly observing
learning and regulation could unravel when, how, and to what extent metacognitive reg-
ulation skills are adopted by RPT participants and consequently help to explain the
potential metacognitive effectiveness of RPT suggested in the current study.

Conclusion

Since the promotion of metacognitive regulation requires explicit modelling and guided
practice, increasing student–staff ratio’s challenge university instructors to successfully

123
High Educ (2015) 70:469–486 485

support students’ regulation (Topping 1996). The present study demonstrated that organ-
ising RPT could be a valuable alternative. A RPT setting concerns a small-scale learning
environment and consequently allows for intensive metacognitive modelling by peers and
individualised feedback on internalised regulation skills. The results of the present study
moreover demonstrated that RPT has the potential to foster university students’ meta-
cognitive regulation, more specifically students’ monitoring. Despite students’ additional
enhanced evaluation and orientation, RPT was considerably less influential towards these
regulation skills. The effects regarding students’ adoption of deep-level regulation were
rather limited as well, given that RPT mainly elicited deep-level comprehension moni-
toring. Based on our findings, we recommend the implementation of RPT in higher edu-
cation in order to promote students’ metacognitive regulation, more specifically their
adoption of (deep-level) monitoring strategies. Additionally, we advise instructors to
carefully train students for RPT and to design appropriate learning materials which can
elicit regulative acts.
The present study not only has the potential to inspire educators when aiming to foster
students’ metacognitive regulation, and it also offers empirical insights enhancing our
understanding of students’ adoption of particular regulation skills. Our findings further-
more raise questions concerning which elements in the RPT setting specifically evoke
metacognitive regulation and therefore present new process-oriented research directions to
explore regulation in collaborative groups in depth.

References

Brown, A. L. (1987). Metacognition, executive control, self-regulation and other more mysterious mech-
anisms. In F. E. Weinert & R. H. Kluwe (Eds.), Metacognition, motivation and understanding (pp.
65–116). Hillsdale: Laurence Erlbaum Associates.
Cheng, Y. C., & Ku, H. Y. (2009). An investigation of the effects of reciprocal peer tutoring. Computers in
Human Behavior, 25, 40–49.
Chi, M., Siler, S., Jeong, H., Yamauchi, T., & Hausmann, R. (2001). Learning from human tutoring.
Cognitive Science, 25, 471–533.
De Backer, L., Van Keer, H., & Valcke, M. (2012). Exploring the potential impact of reciprocal peer
tutoring on higher education students’ metacognitive knowledge and regulation. Instructional Science,
40, 559–588.
Chin, C., & Brown, D. (2000). Learning in Science: A comparison of deep and surface approaches. Journal
of Research in Science Teaching, 37, 109–138.
Duran, D., & Monereo, C. (2005). Styles and sequences of collaborative learning in fixed and reciprocal peer
tutoring. Learning and Instruction, 15, 179–199.
Falchikov, N. (2001). Learning together. Peer tutoring in higher education. London: Routledge Falmer.
Greene, J. A., & Azevedo, R. (2009). A macro-level analysis of SRL processes and their relations to the
acquisition of sophisticated mental models. Contemporary Educational Psychology, 34, 18–29.
Greene, J. A., Robertson, J., & Croker Costa, L. A. (2011). Assessing self-regulated learning using think-
aloud methods. In B. J. Zimmerman & D. H. Schunk (Eds.), Handbook of self-regulation of learning
and performance (pp. 313–328). New York: Routledge.
King, A. (1997). Ask to think-tell why: A model to transactive peer tutoring for scaffolding higher level
complex learning. Educational Psychologist, 32, 221–235.
Meijer, J., Veenman, M. V. J., & van Hout-Wolters, B. H. A. M. (2006). Metacognitive activities in text-
studying and problem-solving: Development of a taxonomy. Educational Research and Evaluation, 12,
209–237.
Moos, D. C., & Azevedo, R. (2009). Self-efficacy and prior domain knowledge: To what extent does
monitoring mediate their relationship with hypermedia learning? Metacognition and Learning, 4,
197–216.
Nota, L., Soresi, S., & Zimmerman, B. J. (2004). Self-regulation and academic achievement: A longitudinal
study. International Journal of Educational Research, 41, 198–215.

123
486 High Educ (2015) 70:469–486

Perry, N. E., & Winne, P. H. (2013). Tracing students’ regulation of learning in complex collaborative tasks.
In S. Volet & M. Vauras (Eds.), Interpersonal regulation of learning and motivation: Methodological
advances (pp. 45–66). London: Routledge.
Pintrich, P. R. (2004). A conceptual framework for assessing motivation and self-regulated learning in
college students. Educational Psychology Review, 16, 385–407.
Roscoe, R. D., & Chi, M. (2008). Tutor learning: The role of explaining and responding to questions.
Instructional Science, 36, 321–350.
Schunk, D. H., & Zimmerman, B. J. (2007). Influencing children’s self-efficacy and self-regulation of
reading and writing through modelling. Reading and Writing Quarterly, 23, 7–25.
Topping, K. J. (1996). Effective peer tutoring in further and higher education: A typology and review of the
literature. Higher Education, 32, 321–345.
Veenman, M. V. J., Elshout, J. J., & Meijer, J. (1997). The generality vs. domain-specificity of metacog-
nitive skills in novice learning across domains. Learning and Instruction, 7, 187–209.
Veenman, M. V. J., Kok, R., & Blöte, A. W. (2005). The relation between intellectual and metacognitive
skills in early adolescence. Instructional Science, 33, 193–211.
Volet, S., Vauras, M., & Salonen, P. (2009). Self- and social regulation in learning contexts: An integrative
perspective. Educational Psychologist, 44, 215–226.
Winne, P. H., & Hadwin, A. F. (1998). Studying as self-regulated learning. In D. J. Hacker, J. Dunlosky, &
A. C. Graesser (Eds.), Metacognition in educational theory and practice (pp. 279–306). Hilsdale:
Erlbaum.
Zimmerman & Schunk. (2011). Self-regulated learning and performance: An introduction and an overview.
In B. J. Zimmerman & D. H. Schunk (Eds.), Handbook of self-regulation of learning and performance
(pp. 1–12). New York: Routledge.

123
Copyright of Higher Education is the property of Springer Science & Business Media B.V.
and its content may not be copied or emailed to multiple sites or posted to a listserv without
the copyright holder's express written permission. However, users may print, download, or
email articles for individual use.