Uploaded by Alex Ang

I can assess myself Singaporean primary students

advertisement
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/277951473
I can assess myself: Singaporean primary students' and teachers' perceptions
of students' self-assessment ability
Article in Education 3-13 · January 2016
DOI: 10.1080/03004279.2014.982672
CITATIONS
READS
8
291
1 author:
Hwei Ming Wong
National Institute of Education (NIE), Singapore
6 PUBLICATIONS 34 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Implementation of Self-assessment and Investigation of Feedback in Lower Primary Classrooms View project
A Teacher-Led Interpretation of the Teacher Growth Model: Inquiry into Professional Identity of Singapore Teachers View project
All content following this page was uploaded by Hwei Ming Wong on 01 November 2015.
The user has requested enhancement of the downloaded file.
This article was downloaded by: [117.169.1.85]
On: 10 July 2015, At: 08:52
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: 5 Howick Place, London, SW1P 1WG
Education 3-13: International Journal
of Primary, Elementary and Early Years
Education
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/rett20
I can assess myself: Singaporean
primary students' and teachers'
perceptions of students' selfassessment ability
Hwei Ming Wong
a
a
Click for updates
Centre for Research in Pedagogy and Practice, National Institute
of Education, 1 Nanyang Walk, Singapore 637616, Singapore
Published online: 24 Nov 2014.
To cite this article: Hwei Ming Wong (2014): I can assess myself: Singaporean primary students' and
teachers' perceptions of students' self-assessment ability, Education 3-13: International Journal of
Primary, Elementary and Early Years Education, DOI: 10.1080/03004279.2014.982672
To link to this article: http://dx.doi.org/10.1080/03004279.2014.982672
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions
Education 3–13, 2014
http://dx.doi.org/10.1080/03004279.2014.982672
I can assess myself: Singaporean primary students’ and teachers’
perceptions of students’ self-assessment ability
Hwei Ming Wong*
Centre for Research in Pedagogy and Practice, National Institute of Education, 1 Nanyang Walk,
Singapore 637616, Singapore
Downloaded by [117.169.1.85] at 08:52 10 July 2015
(Received 31 May 2014; accepted 27 October 2014)
Student self-assessment engages the students in purposeful reflection about what they
are learning and how they are learning it. This study investigated the perceptions of
students and teachers towards the students’ self-assessment ability in two Singapore
primary schools. A total of 75 students were taught how to use self-assessment.
Eighteen students’ self-assessments were randomly selected and compared with an
independent panel of teachers’ assessment. The results revealed both differences as
well as similarities between the students’ and teachers’ perceptions of students’ selfassessment ability. The findings and the implications for students are discussed in
concluding the article.
Keywords: self-assessment; primary school; Singapore; students; teachers; formative
assessment
Introduction
Assessment in Singapore has generally been summative in nature, with the purpose of
grading student performance, ranking students against external norms and each other, creating comparisons between schools, as well as opening access to schools and universities
internationally (Tan, Chow, and Goh 2008). Such high-stakes examinations have been
one of the defining pillars of the Singapore education system since its post-independent
days in the 1960s (Tan, Chow, and Goh 2008), creating a strong examination culture in
Singapore (Cheah 1998) whereby teachers become overly focused on teaching to the test.
In this rapidly changing and globalising world, the Ministry of Education (MOE) is concerned that the Singapore education system is becoming too examination-oriented, with
little room for creativity and critical thinking (Lim 2014). As a result, the MOE has been
changing its assessment mode to include alternative modes of assessments (2002, 2008,
2009). Multiple modes and alternative assessments are currently being incorporated into
Singaporean students’ academic assignments, with students’ academic self-assessment
being increasingly introduced into classroom assessments through initiatives such as the
Strategies for Active and Independent Learning (MOE 2004) which uses assessment
tools such as checklists, observation sheets, and rubrics to evaluate students’ level of
attainment.
*Email: hweiming.wong@nie.edu.sg
© 2014 ASPE
Downloaded by [117.169.1.85] at 08:52 10 July 2015
2
H.M. Wong
This article examined the perceptions of students and teachers towards the students’
ability to use self-assessment in two primary schools in Singapore. A total of 75 students
were taught how to use self-assessment with their student work. Eighteen of the 75 students
were randomly selected for the collection of their self-assessment worksheets and corresponding students’ work. After the intervention, all 75 students answered some openended questions about their self-assessment experiences. The 18 students’ self-assessments
were compared with an independent panel of teachers’ assessments. The results revealed
that there were both differences as well as similarities between the students’ and teachers’
perceptions of students’ ability to use self-assessment. The 18 students were also
interviewed.
The article is organised as follows: Section 1 presents the definition of self-assessment,
a brief literature review of students’ ability to use self-assessment, the research objectives,
and the significance of this study. Section 2 describes the research design, the sampling, the
instruments’ design, and the data collection procedures. The results of the students’ and
teachers’ scoring, the self-report, and interviews are shown in Section 3 with Section 4
presenting the discussion and Section 5 concludes the article.
Self-assessment
Student self-assessment, based on certain standards and criteria, engages the students in
deliberate thought and reflection about what they are learning and how they are learning
it. In simple terms, self-assessment ‘involves students in thinking about the quality of
their own work, rather than relying on their teacher as the sole source of evaluative judgements’ (Andrade and Valtcheva 2009, 13). Learning logs, journals, checklists and charts,
learning contracts, and portfolios are some tools of self-assessment.
According to Trunnel (1992), some of the major assumptions behind self-assessment
are that learners can evaluate themselves, that they have specific goals, and that a
process of self-evaluation will make them more interested, encouraged, and aware of
their own learning process. Brown and Glasner (1999) believed that self-assessment
helps to develop in students the ability to think for themselves, or to develop the confidence
in their learning ability and in their ability to evaluate what they learn and continue to learn
when their schooling days are over. For students, self-assessment is a move away from
reliance on others’ judgements, usually teachers, towards a greater dependence on the
ability to make decisions about their own learning for themselves. It encourages students
to be actively involved, motivated, and more responsible for their own learning.
Research in self-assessment: students’ self-assessment ability
There have been studies in self-assessment which investigated students’ self-assessment
ability at different educational levels.
Cassidy (2007) investigated students in higher education and their ability to self-assess.
The undergraduate students’ estimated marks were compared with tutor marks. The findings revealed a tendency for students to underestimate their marks, and greater levels of
self-assessment accuracy were connected to students with higher tutor marks. Likewise,
Higgins, Harris, and Kuehn (1994) also investigated first-grade and second-grade students’
self-assessment of their integrated projects. When the students’ scores were compared to the
teacher’s on the self-assessment ratings, it was found that more than half the scores were in
complete agreement. The teacher rated the students higher than the students did when
there were differences in the scores. Patri (2002) examined the influence of peer feedback
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Education 3–13
3
on self- and peer-assessment of individual oral presentations of university students. The
results illustrated that with clearly set assessment criteria, peer feedback enabled the students to make judgements of their peers comparable to those of the teacher but the students
were not able to judge themselves in a manner similar to the teacher.
Ross, Rolheiser, and Hogaboam-Gray (1998) analysed the effects of self-evaluation on
Grade 5 and 6 students’ performance in Mathematics. The results revealed that the students
became more accurate when they were taught how to evaluate their work, and the students’
tendency to inflate their self-evaluations waned in the treatment group while remaining
unchanged in the control group. Sullivan and Hall (1997) introduced self-assessment to
third year university students as a strategy for monitoring performance and improving learning. The results showed that most students found self-assessment useful for reflecting on
their work and also indicated a good level of agreement between students and lecturer,
but there was an inclination for more students to overestimate than to underestimate their
grades.
Leach (2012) offered 472 university students the opportunity to self-assess with only
38% or 120 students choosing to self-assess themselves with a grade. The results
showed that the students were able to self-assess themselves with grades similar to that
given by the teacher. However, the findings also indicated that there were occurrences of
overestimating and underestimating by the students with the higher achieving students
underestimating their grades and the lower achieving students overestimating. Lew,
Alwis, and Schmidt (2010) examined the accuracy of 3588 first year polytechnic students’
self-assessment ability. The students’ self-assessments were compared with the tutor’s
scores. The findings indicated an ability effect, in that students who were more
academically able were able to assess with higher accuracy than the students who were
less academically able.
Blatchford (1997) studied the accuracy and stability of academic self-assessment of students from 7 to 16 years and the effects of domain/school subject and social comparison
group on developments over time. Students’ assessments of their ability were found to
be important in their school progress. By 11 years, students were more modest in their
assessment of their attainments compared to when they were 7 years old. The results
also showed that the overall lowering of self-assessment with age is accompanied from
11 years, by a growing accuracy in assessing one’s own attainments, and there was more
stability in ratings from 11 to 16 years. Brown (2008) outlined for teachers a practical
approach using the Quick Check strategy, a self-assessment, to increase student engagement
and confidence. When the Quick Check self-assessment was complemented with the four
purposeful teaching behaviours (set student goals, plan progress towards student goals,
record student progress, celebrate student achievement and effort), the students recognised
the improvement in the volume of work completed and in the number of correct responses,
their judgement of their work improved over time, and there was increased confidence and
ownership of learning and achievement.
In summary, research revealed mixed results with regard to students’ ability to selfassess, with students both overestimating as well as underestimating their self-assessment
scores. A number of issues surfaced: there was more research in students’ self-assessment
ability in higher education and much less in primary school education; most research was
conducted in the West and markedly less so in Asia and in Singapore; and most research
was also lacking in interview data which might have provided a richer analysis of the students’ self-assessment ability from the teachers’ and students’ perspectives. More research
is therefore needed on younger primary school students’ self-assessment ability and a need
to triangulate the results from quantitative and qualitative data sources.
4
H.M. Wong
Research objectives
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Based on Wong (2012), this study examined the primary school students’ self-assessment
ability, in terms of the perceptions of primary school students and teachers on the students’
ability to use self-assessment with the following research question:
What are the differences between the Primary 4 (P4) students’ and teachers’ perceptions
of students’ ability to self-assess?
The P4 students (aged 10 years old) were chosen as subjects for this study because they
were deemed to have adequate literacy and language skills. Lower primary students such as
those in Primary 1, 2, and 3 (aged 7, 8, and 9 years old) might not have the language skills to
participate in the study. Upper level students such as those in Primary 5 and 6 were not
involved because they had a tight curriculum schedule, and Primary 6 students were preparing for the high-stakes national examination, Primary School Leaving Examination at the
end of the year.1
Significance of the current study
This study hopes to bring about empirical evidence of how the understanding of primary
school students’ self-assessment ability is of significance and can potentially contribute
to the use of self-assessment in the classroom from an Asian perspective. The students’
self-assessment ability is examined through the comparison between teachers’ assessment
and the students’ own self-assessment as well as students’ self-report because they provide
an important source of data and feedback about students’ self-assessment ability. How they
perceive their ability to self-assess will have an impact on how self-assessment will be used
by them in the long term. Given that students’ perception of their self-assessment abilities
will have a lasting effect on how they view assessment and learning, the focus on primary
school students is timely because students playing an active role in the assessment process,
such as self-assessing, is an essential progression in students becoming active participants
in the learning processes (Dann 2002). This study therefore aims to contribute to the international literature on how young students’ view of their ability to self-assess can result in
them being more aware of and receptive to the use of self-assessment.
Methodology
Sampling
The sample of this study was selected based on purposive sampling method. The two mixed
gender ‘neighbourhood’ schools, school A and school B, were chosen based on their
average ranking in the national school league table. They were matched in terms of
student characteristics (i.e. academic achievements) and student–teacher ratio. The
‘neighbourhood schools’ [are] so named because they are located in mostly underprivileged,
working-class neighbourhoods and receive students from these areas. With few exceptions,
the neighbourhood schools tend to be poorly positioned in the national school league table,
a ranking and appraisal system that bands schools annually according to academic performance. (Kramer-Dahl and Kwek 2011, 162)
In each school, one class of P4 students was trained in the use of self-assessment. In school
A, class A was the intervention class and its teacher, teacher A, was the intervention teacher.
Likewise, in school B, class B was the intervention class and its teacher, teacher B, was the
intervention teacher.
Education 3–13
5
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Participants
The study started out with 80 P4 students (38 and 42 students from classes A and B, respectively) and 2 teachers. At the end of the study, due to absenteeism from illnesses and transfer
outs, a total of 75 P4 students participated (37 and 38 students from classes A and B,
respectively).
From both intervention classes, a subgroup of the intervention students were randomly
selected by their teachers for the collection of their self-assessment worksheets and corresponding students’ work. Each intervention teacher randomly selected nine students from
her class – three students with high ability, three students with medium ability, and three
students with low ability. Therefore, a total of 18 intervention students were selected for
further analysis of their self-assessment and corresponding student work and their selfassessment experiences.
Two teachers, Teacher X and Teacher Y, were recruited to score the 18 students’ work,
together with the researcher. Both Teacher X and Teacher Y were current primary school
teachers with over six years of teaching experience and had experience in using rubrics
to score students’ work. The researcher (Teacher Z) previously taught for six years and
also had experience with rubrics design.
Subject
The investigation is focussed only on Mathematics, which is largely a procedural subject.
As a result, it is easier for the primary school students to see their own progress from ‘not
understanding’ to ‘understanding’, as compared to the other subjects such as English
language and Science. Another reason for investigating self-assessment in Mathematics
was that both schools were using the same Mathematics textbooks and workbooks (there
were three known approved Mathematics textbooks and workbooks by MOE that the
schools could choose from) which would constitute a fair comparison between the two
intervention classes. During data collection, the Mathematics topics taught in both
schools included Angles, Perpendicular Lines, Fractions, Decimals, and Measures.
Instruments and procedures
Student self-assessment strategy worksheets
A set of student self-assessment strategy worksheets was created for the intervention:
checklist, learning log, and rubrics. The self-assessment strategies used the following criteria: (1) Deep understanding – the students’ understanding of the problem or task, (2) Strategies and reasoning – the students’ ability to solve the problem or task through strategies
and reasoning, (3) Clarity – the clarity of their solution, (4) Written communication –
how they conveyed their solutions, and (5) Effort – the extent of effort they put into
their work.
The checklist worksheet required the students to read two sentences for each criterion
and to select the relevant boxes – ‘Yes’, ‘No’, or ‘Unsure’ – to reflect their thinking of their
own work, for example, ‘I understand what the questions want.’ (understanding), and ‘I try
my best to solve the questions.’ (effort).
In the learning log worksheet, for each of the five criteria, the students were required to
write what they thought or felt about their work; for example, ‘What do I think about my
understanding? Good or Bad? Why?’ (understanding) and ‘What methods did I use to solve
the questions?’ (strategies and reasoning).
6
H.M. Wong
Downloaded by [117.169.1.85] at 08:52 10 July 2015
For the rubrics worksheet, the students assessed themselves by selecting one of the four
levels they were at for each of the five criteria. The students had to read the descriptors for
the four different levels and select the level most suitable for themselves for each of the five
criteria, for example, ‘I try but I don’t understand what the questions want.’ (understanding,
level 1) and ‘I have very good understanding of what the questions want.’ (understanding,
level 4).
The students were given instructions on how to use each of the three self-assessment
strategies to assess their work. They were then given two weeks or a fortnight to use the
self-assessment strategy before the next one was introduced and taught. The students
used one self-assessment strategy three times in each fortnight. The students had to use
self-assessment on a regular basis for seven fortnights.
Student work scoring rubrics and exemplars for use by teachers
Another set of rubrics was developed to score the students’ work and student work exemplars were also written. The rubrics focussed on the following in which the teachers based
their assessment of the students’ work on: (1) deep understanding, (2) strategies and reasoning, (3) clarity, and (4) written communication by indicating one of four levels, for example,
‘Uses wrong/no strategies for problem-solving’ (strategies and reasoning, level 1) and
‘Uses appropriate strategies for all problem-solving’ (strategies and reasoning, level 4).
Exemplars of the student work demonstrated and explained what each level of the scores
meant and to calibrate the teachers’ scoring during the training. The researcher (Teacher
Z) met with Teachers X and Y for training in the general rules of scoring and the explanation of each of the criteria in the rubrics. The teachers tried out their scorings of
student work examples and discussed their score discrepancies to reach common understandings. The actual scoring of the intervention students’ work began when the teachers
were fairly consistent in their scorings with each other, at least 75% and above of exact
agreement between them.
The student works were divided into high, medium, and low based on students’ abilities. The teachers were not informed of the division of student ability so that they would
maintain neutrality and objectivity. Teachers X, Y, and Z each scored the student work independently and then proceeded to compare and justify their scores. Discussions were carried
out to reach consensus on a final score when there were discrepancies in the scores.
Through the training of scoring student work samples and the discussions to reach consensus scores, the extent of agreement between the teachers was improved and thus the interrater reliability (IRR) was also enhanced. At each time, two teachers scored the same
student works from one ability group and each teacher scored student works from two
ability groups. See Table 1.
Table 1. Student work scoring by teachers.
Student work ability type
High ability
Medium ability
Low ability
Teachers who are scoring
X
Z (researcher)
X
Y
Y
Z (researcher)
Education 3–13
7
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Student self-report and interview
After four months of intervention (April, May, July, and August), the 75 intervention students completed a self-report on questions such as ‘Do you think you have assessed yourself fairly in your own work?’, and ‘Do you need more practice to be confident and
experienced to assess your own work?’
From the 75 students, a subset of 18 students, whose self-assessments and student
works were collected, was also interviewed after the self-report. The interview questions
were constructed beforehand and each student was asked the same questions so that
there would be comparability of the students’ responses. The students were asked questions
such as ‘Do you think students can assess themselves or only teachers can assess students?
Why?’ The questions sought to conduct an in-depth examination of the students’ selfassessment experiences. The group interview format was used with the students to minimise
disruptions to the students’ lessons. Being in a group situation, the students might also be
less inhibited and be forthcoming in giving their responses (Alder and Clark 2008). From
each intervention class, the students were interviewed in groups of three according to their
ability grouping; for example, the three students with high ability were interviewed together
as a group so that the group would be small enough to allow everyone sufficient time to
speak.
Teacher interview
After the intervention, a semi-structured interview was conducted with the two teachers
separately to gather further information of their perceptions of self-assessment in Singapore
classrooms. Questions such as ‘Do you think students are capable of assessing themselves
fairly?’ and ‘What do you think are required of the students before they can assess themselves fairly?’ were asked.
Analysis and results
Student work
The students’ self-assessment worksheets and related work were collected from 18 students.
These included checklists, learning logs, and rubrics that were conducted three times over a
fortnight for a period of four months (or seven fortnights). Each student had to assess themselves on a piece of their schoolwork, using a self-assessment worksheet. The students were
divided into high-, medium- and low-ability based on their teachers’ judgements of their
achievement in daily Mathematics classwork, generating a total of six high-ability students,
six medium-ability students, and six low-ability students. A total of 122, 126, and 120 selfassessment worksheets and related student works were collected from the high-ability,
medium-ability, and low-ability students, respectively, at the end of 7 fortnights. Due to
illness and the students being absent, there was a difference in the number of self-assessment worksheets and related work collected from the different ability students. Table 2 presents the number of self-assessment worksheets and student work collected from the 18
students from the two intervention classes, A and B.
Two teachers and the researcher participated in the scoring of the students’ work. Each
teacher scored the students’ works from two different ability groups. At each time, two
different teachers scored the same ability group’s student works individually and then compared their scores. When there was a discrepancy between scores, the teachers would
discuss to reach a consensus score.
8
H.M. Wong
Table 2. Collection of 18 students’ self-assessment worksheets and student work.
Ability
Time
Type SA
Class
High
Fortnight 1
Checklist
Fortnight 2
Learning log
Fortnight 3
Rubrics
A
B
A
B
A
B
A
B
A
B
A
B
A
B
9
9
9
8
8
8
8
9
9
9
9
9
9
9
122
Fortnight 4 Refresher: checklist, learning log, rubrics
Fortnight 5 Students’ choice: rubrics (15), checklist (3)
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Fortnight 6 Students’ choice: rubrics (14), checklist (4)
Fortnight 7 Students’ choice: checklist (13) rubrics (4), learning
log (1)
Total SA collected a
Medium Low
9
9
9
9
9
9
9
9
9
9
9
9
9
9
126
9
9
9
7
9
7
7
9
9
9
9
9
9
9
120
a
The number of SA collected differed because there were four, zero, and six absentees, respectively, for high-,
medium- and low-ability student work.
SA: self-assessment.
Teachers’ IRR
The teachers’ IRR were estimated using the percentage of exact agreement and Cohen’s
kappa coefficient for the domains of understanding, reasoning, communication, and
clarity. The teachers’ IRR were conducted as a means of verifying coherence in the
teachers’ understanding of the student work scoring criteria and to increase the credibility
of the results. Table 3 summarises the teachers’ IRR using Cohen’s kappa coefficient.
Table 3. Teachers’ IRR in scoring high-ability, medium-ability, and low-ability student work using
Cohen’s kappa coefficient.
Student work
High ability (n = 122)c
Medium ability (n = 126)c
Low ability (n = 120)c
Domain
Understanding
Reasoning
Communication
Clarity
Understanding
Reasoning
Communication
Clarity
Understanding
Reasoning
Communication
Clarity
Asymp.
Kappa value std. errora Approx. Tb Approx. sig.
.894
.952
.913
.881
.690
.475
.412
.357
.815
.789
.837
.826
.039
.034
.059
.065
.061
.087
.119
.136
.044
.047
.044
.050
12.325
12.714
11.757
11.254
9.713
6.769
5.462
5.052
13.082
12.654
12.199
10.065
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
Note: IRR refers to the extent to which independent evaluators produce similar ratings in judging the same abilities
or characteristics in the same target person or object (American Psychological Association, 2014).
a
Not assuming the null hypothesis.
b
Using the asymptotic standard error assuming the null hypothesis.
c
n for high-, medium-, and low-ability student work are different due to four, zero, and six absentees, respectively.
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Education 3–13
9
The percentage of exact agreement by Teachers X and Z for high-ability student work
was 94.26% for understanding; 98.36% for reasoning; 98.36% for communication; and
97.54% for clarity. The IRR were high for each of the domains. According to research,
when the percentage of exact agreement is 70% and above, the IRR are considered to be
good. The kappa values were .89 for understanding, .95 for reasoning, .91 for communication, and .88 for clarity (Table 3). As a rule of thumb, values of kappa from .21 to .40
are considered fair, .41 to .60 moderate, .61 to .80 substantial, and above .80 outstanding
(Landis and Koch 1977). The kappa values were therefore considered significant.
The percentage of exact agreement by Teachers X and Y for medium-ability student
work was 83.33% for understanding, 84.13% for reasoning, 91.27% for communication,
and 91.27% for clarity. The kappa values were .69 for understanding, .48 for reasoning,
.41 for communication, and .36 for clarity. The kappa values were considered significant
for understanding, and moderate for reasoning and communication. However, the kappa
value for clarity was fair (Table 3). Although the IRR were high, the kappa values for
the medium-ability student work were moderate for reasoning and communication, and
only fair for clarity. This may be due to the way teachers evaluated medium-ability
student work, whereby they oscillated between two different levels (e.g. Levels 2 and 3
of the rubrics). This was different from high- or low-ability student work that was more
clearly distinguishable between the different levels for the teachers. Furthermore, it
would appear that teacher personality might have played a role in the low kappa values.
For example, some teachers differed in their tolerance for clarity in student work as exhibited by lack of coherence, poor handwriting, and so on.
The percentage of exact agreement by Teachers Y and Z for low-ability student work
was 87.50% for understanding, 85.83% for reasoning, 90% for communication, and
90.83% for clarity. Again, the IRR found were high. The kappa values were .82 for understanding, .79 for reasoning, .84 for communication, and .83 for clarity (Table 3).
Teachers’ scoring
Table 4 presents the mean scores and standard deviations of teachers’ scoring for high-,
medium-, and low-ability student work. Teachers X and Z’s mean scores when scoring
122 pieces of high-ability student works in terms of understanding, reasoning, communication, and clarity showed that they were consistent in their scoring, having similar
scores. Their mean score differences ranged from 0.01 to 0.03. The SDs showed no
large variations of the mean scores.
When scoring 126 pieces of medium-ability student works in the four domains, Teachers X and Y’s mean scores showed that they were fairly consistent in their scoring.
Their mean differences ranged from 0.03 to 0.11. Likewise, the means for Teachers Y
and Z when scoring 120 pieces of low-ability student work showed consistency in their
scoring. Their mean differences ranged from 0.04 to 0.08.
Teachers’ versus students’ scoring
The teachers scored the students’ work that the students themselves also scored using the
checklists, learning logs, and rubrics. However, the teachers’ scoring, which was done
using rubrics, could only be compared to the students’ self-assessment rubrics scoring.
The checklists and learning logs were scored in terms of selecting the most appropriate
box of ‘Yes’, ‘No’, and ‘Unsure’ and writing about their feelings on their understanding,
reasoning, communication, clarity, and effort. It was, therefore, unsuitable to compare
10
H.M. Wong
Table 4. Mean scores and standard deviations of teachers’ scoring for high-ability, medium-ability,
and low-ability student work.
Domain
Mean (SD)
Teacher X
Mean (SD)
Teacher Z
High ability (n = 122)a
Understanding
Reasoning
Communication
Clarity
3.48 (.67)
3.75 (.54)
3.88 (.40)
3.87 (.41)
3.51 (.66)
3.74 (.54)
3.88 (.38)
3.86 (.39)
Medium ability (n = 126)a
Understanding
Reasoning
Communication
Clarity
Teacher X
3.46 (.65)
3.78 (.52)
3.88 (.39)
3.89 (.34)
Teacher Y
3.57 (.64)
3.75 (.58)
3.93 (.29)
3.95 (.25)
Low ability (n = 120)a
Understanding
Reasoning
Communication
Clarity
Teacher Y
3.09 (.82)
3.09 (.84)
3.36 (.72)
3.54 (.56)
Teacher Z
3.02 (.87)
3.05 (.84)
3.28 (.71)
3.48 (.58)
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Student work
a
n for high-, medium-, and low-ability student work are different due to four, zero, and six absentees, respectively.
the students’ responses in the checklist and learning log with that of the teachers’ scoring of
the same related student work. As such, there were a total of 55, 54, and 55 pieces of student
work (in which the students assessed themselves using rubrics) from the high-ability,
medium-ability, and the low-ability students, respectively. Table 5 shows both the teachers’
scores and the different ability students’ scores of the related student work using rubrics in
terms of understanding, reasoning, communication, and clarity.
Regardless of ability, the students’ mean scores were lower than the teachers’ consensus
mean scores in all four domains of understanding, reasoning, communication, and clarity.
For the high-ability students, their mean scores were lower than the teachers’ scores in
communication and clarity with the biggest mean differences of 0.62, followed by reasoning (0.45) and understanding (0.27). It suggested that the high-ability students might not be
as confident in their communication and clarity. For the medium-ability students, their communication scores had the biggest mean difference of 0.43 when compared to the teachers’
Table 5. Teachers’ and the different ability students’ mean scores in understanding, reasoning,
communication, and clarity.
High ability (n = 55)a
Understanding
Reasoning
Communication
Clarity
Medium ability (n = 54)a
Low ability (n = 55)a
Student
selfassessment
M (SD)
Teacher
consensus
M (SD)
Student
selfassessment
M (SD)
Teacher
consensus
M (SD)
Student
selfassessment
M (SD)
Teacher
consensus
M (SD)
3.22 (.57)
3.35 (.62)
3.25 (.55)
3.25 (.67)
3.49
3.80
3.87
3.87
3.43 (.60)
3.43 (.57)
3.50 (.54)
3.63 (.49)
3.67 (.55)
3.80 (.49)
3.93 (.26)
3.94 (.23)
2.85 (.65)
2.84 (.60)
3.07 (.63)
3.13 (.61)
3.13
3.22
3.40
3.49
(.69)
(.49)
(.39)
(.39)
a
The difference in n is due to more high- and low-ability students selecting rubrics as their choice of selfassessment in Fortnights 5, 6, and 7 (see Table 2).
(.82)
(.81)
(.66)
(.54)
Education 3–13
11
Table 6. Independent sample t-test of teachers’ and the different ability students’ scores in
understanding, reasoning, communication, and clarity.
Student
work
High ability
Medium
ability
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Low ability
Domain
t
df
Sig. (2tailed)
Mean
difference
Std. error
difference
Understanding
Reasoning
Communication
Clarity
Understanding
Reasoning
Communication
Clarity
Understanding
Reasoning
Communication
Clarity
−2.263
−4.297
−6.800
−5.905
−2.171
−3.620
−5.200
−4.288
−1.936
−2.808
−2.661
−3.310
104.085
102.585
96.853
86.284
106
103.718
76.960
75.701
102.788
99.689
107.881
108
.026
.000
.000
.000
.032
.000
.000
.000
.056
.006
.009
.001
−.273
−.455
−.618
−.618
−.241
−.370
−.426
−.315
−.273
−.382
−.327
−.364
.121
.106
.091
.105
.111
.102
.082
.073
.141
.136
.123
.110
scores, followed by reasoning (0.37), clarity (0.31), and understanding (0.24). The mediumability students seemed to be least confident about communication. The low-ability students’ scores were lower than the teachers’ scores in all four domains. Reasoning had the
biggest mean difference of 0.38, followed by clarity (0.36), communication (0.33), and
understanding (0.28).
Independent sample t-tests were conducted to further examine the difference between
the different ability group students’ and the teachers’ scores. Table 6 presents the differences
between the students’ and the teachers’ scores for understanding, reasoning, communication, and clarity for the high-ability, medium-ability, and low-ability students.
The results revealed that there were significant differences between the teachers’ and
students’ scores in understanding, reasoning, communication, and clarity in the highability and medium-ability students, although there was smaller difference between the students and the teachers in terms of understanding. It suggested that the students’ scores were
more in agreement with the teachers’ scores in terms of understanding as compared to the
other three domains. There were significant differences between the teachers’ and students’
scores in reasoning, communication, and clarity but not in understanding for the low-ability
students. It appeared that the low-ability students were in agreement with the teachers’
scoring for the domain of understanding but not for the other three domains of reasoning,
communication, and clarity. In general, the teachers scored the students’ work more positively than the students themselves. The students seemed to have been more modest than
necessary and depressed their own scores when assessing themselves. This might be indicative of students needing more practice in self-assessment in order to give a more accurate
picture of their work and knowledge.
Students’ self-report and interview responses
After the intervention, the students completed a self-report about their self-assessment
experiences. Questions such as ‘Do you think students should take part in assessing their
own work?’ and ‘Do you think you have assessed yourself fairly in your own work?’
were asked.
Table 7 shows the 75 intervention students’ responses to specific questions about their
self-assessment ability after they have been through the use of self-assessment.
12
H.M. Wong
Table 7.
Students’ self-report responses about their self-assessment ability.
Frequency
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Students should take part in assessing their own work
No
1
Yes
74
Total
75
I have assessed myself fairly in my own work
No
4
Yes
71
Total
75
I need more practice to be confident to assess my own work
No
8
Yes
66
Missing/NA
1
Total
75
Per cent
1.3
98.7
100.0
5.3
94.7
100.0
10.7
88.0
1.3
100.0
Table 8. Response count of students’ reasons to why students can assess themselves.
Common responses/reasons
1. Have information about self/know ourselves better than the teachers
2. To learn more about ourselves
3. Take ownership for learning
Count (n = 18)
Per cent
9
7
2
50
38.89
11.11
Majority of the students thought that they should take part in assessing their own work
(98.7%). Although 94.7% of the students felt that they have assessed themselves fairly, 88%
also believed that they needed more practice to be confident in their own self-assessment.
One specific question was asked of the 18 students about their self-assessment ability,
‘Do you think students can assess themselves or only teachers can assess students? Why?’
in the interviews. All 18 students reported that students could assess themselves and not
only teachers. A variety of reasons were given for why they could assess themselves
(Table 8). The most frequent reason given by the students was that they had information
about themselves or that they knew themselves better than their teachers.
The following students’ interviews were indicative of what the students felt were the
reasons why they could also assess themselves.
Student M3: (class A)
Student L1: (class A)
Student HC: (class B)
Student MA: (class B)
Yes. I know myself better and can help me to learn more
Yes, when we mark [assess] by ourselves, we can know ourselves better
Yes. Because even if the teacher helps [assesses] you but you don’t want to
study [listen], it would be of no use. But when students assess themselves,
it will be of more use as they want to study
Both [students and teachers can assess students]. The students will have a
chance to assess themselves, to learn more
Teachers’ interview responses
The two intervention teachers were asked if they perceived that their students were capable
of assessing themselves fairly, and what were the requirements to do so. The teachers
believed that their students were capable of assessing themselves but they mentioned
Education 3–13
13
several requirements that needed to be in place before students could assess themselves
fairly, such as knowing how to read, the training on the use of self-assessments such as
rubrics, and the number of times the self-assessment was used.
Teacher A: (class A)
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Teacher B: (class B)
I think, at P4, they can and are capable of assessing themselves, you know.
They must be able to read. (laughs) [They must] Understand the questions,
and what is needed of them to do the questions – the understanding and the
methods. I think training is also important, you know, like training them on
how to use the rubrics so they know how to assess themselves.
Yes. But the thing is that, in fact, I think we should not do this too often
because a lot of the time, let’s say especially we are using the rubrics and
checklists, a lot of the time they are so lazy that they simply write down the
number they think they are [rather than what they actually are]. … I would
think for certain topics, probably you will see a difference in the scoring that
they have put down. But let’s say it is the day-to-day work, then probably it
is almost the same, scoring wise. … They also need to know what they need
to achieve, what they are striving for, if not there’s no purpose [in assessing
themselves]. Of course, they need to be familiar with the rubrics too so that
they know where they are at. Of course, teaching them how to use the
rubrics, and others are important too. If not, they won’t know how to use
[them].
Discussion
When examining the students’ self-assessment ability, the statistical results showed that
there were significant differences between the students’ scoring and the teachers’ scoring
on the same piece of student work except for the low-ability students in terms of understanding. It was interesting to note that the low-ability students’ scores were in agreement
with the teachers’ in terms of understanding. This finding was not consistent with the findings in Lew, Alwis, and Schmidt (2010) where the more academically abled students were
found to be able to assess with greater accuracy. The low-ability students were likely to be
more honest in knowing what they did not know and reflected that in their scoring. In
general, the students, regardless of abilities, tended to score themselves lower on their
work than the teachers did or underestimated their scores as compared to the teachers’
scores. This finding was again not consistent with that found in Leach (2012) where the
higher achieving students were underestimating and the lower achieving students
were overestimating. It is likely that the students were not used to self-assessment and
were not confident in assessing their own work.
The results from the student self-report indicated that the students felt that they should
assess themselves (98.7%) and that they have assessed themselves fairly (94.7%) because,
through their interviews, they believed that they have information about themselves that
they could use that the teachers did not have. At the same time, the majority of the students
(88%) also indicated that they needed more practice to be confident to assess their own
work. The teachers, when interviewed, commented that their students were capable of
assessing themselves but they needed guidance and training in the use of self-assessment.
Despite the conflicting results indicated by the students, the students understood that,
most importantly, they still needed practice in self-assessment to feel confident and be
experienced in order to accurately assess their own work even though they might have
information about their own learning that the teachers did not.
The results implied that the accuracy of the students’ self-assessment would be affected
by their lack of confidence, by the lack of practice and training in the use of self-assessment.
Downloaded by [117.169.1.85] at 08:52 10 July 2015
14
H.M. Wong
This finding coincided with similar conclusions stated in Cassidy (2007), Higgins,
Harris, and Kuehn (1994) and Patri (2002) where their students also underestimated their
self-assessment scores. Concurrently, this finding was inconsistent with that found in
Ross, Rolheiser, and Hogaboam-Gray (1998), and Sullivan and Hall (1997) where their students overestimated their scores instead. However, Blatchford (1997), Brown (2008), and
Ross, Rolheiser, and Hogaboam-Gray (1998) showed that over time, the students tended to
improve in their judgement of their own work. By indicating that they needed more time
and practice in order to self-assess accurately, the students in this study demonstrated
that they had the awareness that they would improve in their judgement of their work
when they had more time and practice in self-assessment.
There were more similarities than differences between the students’ and teachers’ perceptions of students’ ability to use self-assessment. Both the students’ and the teachers’
views were similar and concurred that students could assess themselves. Both the students
and teachers also agreed that training in the use of self-assessment was important and
needed. However, there was a difference between the students and the teachers when
looking at the actual act of scoring their work. The students tended to be tougher on themselves when they were scoring their own work as compared to the teachers’ scoring.
Conclusions
In concluding this article, the following limitations should be noted. In this study, only 2
classes participated in the self-assessment intervention and only 18 students’ self-assessments were compared with the teachers’ assessments. Thus, the results obtained could
not be assumed to be representative of all students in Singapore. For future studies, the
sample size could be increased to involve more classes and students to study the students’
self-assessment ability as well as to study the effect on students’ ability to self-assess.
This intervention was carried out for only four months because the timing was stipulated by MOE when permission was sought to conduct the study in the schools. For
future studies, the length of time for the study should be lengthened, possibly longitudinal,
in order to study if the accuracy of the students’ self-assessment ability would increase over
time and if the students’ ability has an effect on their ability to self-assess over time.
Notwithstanding the limitations, the findings indicated that students as young as 10
years of age have the ability to assess themselves. When given the opportunity to use
self-assessment and with training in self-assessment, the students were able to take ownership for their learning and they were not adverse to using self-assessment. The need for
practice in using self-assessment was highlighted by the students. This indicated that
they were not confident in assessing themselves accurately, and needed more practice in
using self-assessment. A possible reason for the students feeling less than confident to
assess themselves accurately is that teachers have traditionally been the ones to assess
the students, be it conventional or alternative assessment. It would therefore take the students some time to get used to assessing themselves through self-assessment. This selfawareness by the students of needing more practice also implied their receptiveness to
using self-assessment which would potentially contribute towards the regular use of selfassessment in the classrooms.
Predominantly, research in students’ self-assessment ability has been conducted in secondary education and higher education in the West. Based on the results revealed in this
study, it is argued that self-assessment has a place in the primary school classrooms
because the students have the ability to self-assess and need not rely ‘on their teacher as
the sole source of evaluative judgements’ (Andrade and Valtcheva 2009, 13). When
Education 3–13
15
students learn to self-assess more accurately and confidently through practice, they can
move on to learn how to learn, and use the information gained through the self-assessment
process to improve their learning.
Note
1.
The Primary School Leaving Examination (PSLE) is a national examination that is conducted in
Singapore annually, which students have to take at the end of the final year of their six years of
primary school education.
Downloaded by [117.169.1.85] at 08:52 10 July 2015
References
Alder, E. S., and R. Clark. 2008. How it’s Done: An Invitation to Social Research. 3rd ed. Belmont,
CA: Thomson Wadsworth.
American Psychological Association. 2014. APA Dictionary of Statistics and Research Methods.
Washington, DC: APA.
Andrade, H., and A. Valtcheva. 2009. “Promoting Learning and Achievement Through SelfAssessment.” Theory Into Practice 48 (1): 12–19.
Blatchford, P. 1997. “Students’ Self Assessment of Academic Attainment: Accuracy and Stability
from 7 to 16 Years and Influence of Domain and Social Comparison Group.” Educational
Psychology 17 (3): 345–359.
Brown, W. 2008. “Young Children Assess their Learning: The Power of the Quick Check Strategy.”
Young Children 63 (6): 14–20.
Brown, S., and A. Glasner, eds. 1999. Assessment Matters in Higher Education: Choosing and Using
Diverse Approaches. Buckingham: SRHE and Open University Press.
Cassidy, S. 2007. “Assessing ‘Inexperienced’ Students’ Ability to Self-assess: Exploring Links with
Learning Style and Academic Personal Control.” Assessment & Evaluation in Higher Education
32 (3): 313–330.
Cheah, Y. M. 1998. “The Examination Culture and its Impact on Literacy Innovations: The Case of
Singapore.” Language and Education 12 (3): 192–209.
Dann, R. 2002. Promoting Assessment as Learning: Improving the Learning Process. New York:
RoutledgeFalmer.
Higgins, K. M., N. A. Harris, and L. L. Kuehn. 1994. “Placing Assessment into the Hands of Young
Children: A Study of Student-generated Criteria and Self-assessment.” Educational Assessment
2 (4): 309–324.
Kramer-Dahl, A., and D. Kwek. 2011. “Reading the Home and Reading in School: Framing Deficit
Constructions as Learning Difficulties in Singapore English Classrooms.” In Multiple
Perspectives on Difficulties in Learning Literacy and Numeracy, edited by C. Wyatt-Smith, J.
Elkins, and S. Gunn, 159–178. Dordrecht: Springer.
Landis, J. R., and G. G. Koch. 1977. “The Measurement of Observer Agreement for Categorical
Data.” Biometrics 33 (1): 159–174.
Leach, L. 2012. “Optional Self-assessment: Some Tensions and Dilemmas.” Assessment &
Evaluation in Higher Education 37 (2): 137–147.
Lew, M. D. N., W. A. M. Alwis, and H. G. Schmidt. 2010. “Accuracy of Students’ Self-assessment
and their Beliefs about its Utility.” Assessment & Evaluation in Higher Education 35 (2):
135–156.
Lim, L. 2014. “What’s Wrong with Singaporeans?” In Hard Choices: Challenging the Singapore
Consensus, edited by D. Low, and S. Vadaketh, 79–96. Singapore: NUS Press.
Ministry of Education (MOE). 2002. Report of the Junior College/Upper Secondary Education
Review Committee. Singapore: SNP SPrint Pte.
Ministry of Education (MOE). 2004. Strategies for Active and Independent Learning. Singapore:
Curriculum Planning and Development Division, Ministry of Education.
Ministry of Education (MOE). 2008. Speech by Dr Ng Eng Hen, Minister for Education and Second
Minister for Defence, MOE Work Plan Seminar 2008, on Thursday, September 25 2008 at 9:30 a.
m., Ngee Ann Polytechnic Convention Centre. http://www.moe.gov.sg/media/speeches/2008/09/
25/speech-by-dr-ng-eng-hen-at-the-moe-work-plan-seminar-2008.php
16
H.M. Wong
Downloaded by [117.169.1.85] at 08:52 10 July 2015
Ministry of Education (MOE). 2009. Report of the Primary Education Review and Implementation
Committee. http://www.moe.gov.sg/initiatives/peri/files/peri-report.pdf
Patri, M. 2002. “The Influence of Peer Feedback on Self- and Peer-assessment of Oral Skills.”
Language Testing 19 (2): 109–131.
Ross, J. A., C. Rolheiser, and A. Hogaboam-Gray. 1998. “Impact of Self-Evaluation Training on
Mathematics Achievement in a Cooperative Learning Environment.” Paper presented at the
Annual Meeting of the American Educational Research Association, San Diego, CA. (ERIC
Document Reproduction Service No. ED422381), April.
Sullivan, K., and C. Hall. 1997. “Introducing Students to Self-Assessment.” Assessment & Evaluation
in Higher Education 22 (3): 289–305.
Tan, Y. K., H. H. Chow, and C. Goh. 2008. Examinations in Singapore: Change and Continuity
(1891–2007). Singapore: World Scientific.
Trunnel, P. 1992. Self-Assessment. Adventures in Assessment. Vol. 2. Boston, MA: SABES. http://
www.sabes.org/resources/adventures/vol2/2trunnell.htm
Wong, H. M. 2012. The perceptions of Singaporean teachers and students toward academic selfassessments (Unpublished doctoral dissertation). National Institute of Education, Nanyang
Technological University, Singapore.
View publication stats
Download