Role of Self Assessment in Learning

My interest in self-assessment stems from personal experiences of being assessed and the frustration felt when most assessed work was simply awarded a grade, contained minimum feedback if any and was then expected to be filed away despite the many questions I may have had. However, many years later, while attended a language teaching training course, I was given a self-evaluation sheet to complete by the instructor. Uncomfortable as this was, I realised that this was the first time that I had been presented a format to self-assess/reflect on my work. After completing the sheet and the subsequent discussion about the contents, the instructor provided feedback of a type that I could use, in a context which was supportive and which respected my goals as a language teacher.

Since then I have developed an interest in how self-assessment can be used to promote learning in an English as a Foreign Language (EFL) classroom. During my teaching in Asia I have seen the need for learners to take greater responsibility for their own learning in order to move away from the more traditional teacher-led, didactic approach. I have found that using self-assessment as part of reflective learning can lead to greater ownership and autonomous learning as more attention is paid to how learners acquire knowledge.

This essay evaluates the role of assessments, particularly self-assessment as a tool for promoting learning, as I recount the journey taken with my Chinese learners on an English Pathways Program (EPP) and what has led to the decision for using a range of formative tasks contained in a portfolio of written work, with learners ultimately taking ownership of their learning.

What is the role of assessments?

Assessment according to Gipps (1994, p. vii) is: a wide range of methods for evaluating student performance and attainment including formal testing and examinations, practical and oral assessment, and classroom based assessment carried out by teachers and portfolios. Many curricula in language schools reflect Tyler’s (1949) classical model that specified objectives, content, and means of achieving and assessing pre-determined learning outcomes. This model of behaviourism views the learner as a passive absorber of information provided by the teacher and in this way learning becomes an incidental rather than an intentional process. Gipps (1994) argues that the dominance of this model in the classroom, has meant that teachers have focused their instruction on discrete skills and on decontextualized test items, with continued practice until mastery is achieved. Black and Wiliam (1998a) found this type of testing encourages superficial or ‘shallow’ rote learning, as learning isolated facts, quickly disappear from the memory because they have no meaning and do not fit into the learner’s conceptual map. This has been witnessed many time in our classrooms where on one day students are able to recite easily a list of vocabulary or grammatical rules, as they have just done that in class or in a test, but are unable to recall the same information, a few days later. An alternative to this behaviourist/objectives model comes from constructivist psychology which argues that knowledge is not directly transmittable from person to person, but rather is individually constructed or discovered. Glasersfeld (1989) argues that the responsibility of learning should reside increasingly with the learner and constructivism emphasizes the importance of the learner being actively involved in the learning process, unlike previous educational viewpoints where as we noted above the responsibility rested with the instructor to teach and where the learner played a passive, receptive role. Glasersfeld (1989) urges that learners be taught how to learn by engaging their metacognitive functions, resulting in learning being an intentional process and leading to ‘deep’ learning. Sadler (1989) supports this by saying that developments in metacognition tell us students need to become competent assessors of their own work. McDonald and Boud (2003) have argued that “the formal development of self-assessment skills is an important part of the curriculum at all levels” (p. 210) with Black and Wiliam (1998b) stating that self-assessment is “an essential component of formative assessment”. In support of this active learning approach, Gipps (1994) advocates for more frequent and a greater range of assessments, such as essays, performance assessments, small group tasks and projects.

Bould (1991) defines self-assessment as ‘the involvement of students in identifying standards and/or criteria to apply to their work and making judgements about the extent to which they have met these criteria and standards.’ (Boud, 1991, p.5). The latter stage, often called self-grading or self-testing is only one aspect of self-assessment and Bould (1995) warns against an overemphasis on this aspect as it can direct attention away from involving learners in identifying and engaging with criteria, a stage which he says is both difficult and often neglected. In China today, assessments give all the power to the teacher, to make unilateral and final judgements on a student’s work. However, if we want our students to become independent, active learners, then this relationship between student and teacher needs to be changed and by incorporating self-assessment into classroom learning, students as well as teachers acknowledge assessment as a mutual responsibility, and not as the sole responsibility of the teacher (Oscarson, 1989).

Other relevant issues concerning assessments noted by Black and Wiliam (1998a) were: the filling in of records rather than analysing students work to identify their learning needs; and the over-emphasis on awarding marks and grades, often using normative referencing, which encourages competition rather than personal improvement. Many of our learners’ experience of normative referencing in their previous schools led them to believe that they lacked ability and as a result they had lost confidence in their own capacity to learn. Therefore, as a result of this, we adopted a more ipsative approach where learners are more focussed on their own gains rather than other’s grades. In support of this Hounsell et al. (2008) noted that awarding grades often comes at the expense of giving useful advice or feedback, which needs to be integral to the ongoing teaching and learning cycle, as achievement gains from formative assessment are amongst the most substantive of all pedagogical interventions. Ellery (2008, p. 422) elaborates on this by saying that, the opportunities for learning are greatest in formative assignments requiring drafts where students receive feedback and have the occasion to actively engage with the feedback to improve the product in its subsequent draft(s), such as in essays. Gipps (1994) argues for the use of ‘qualitative’ descriptors believing that collapsing or aggregating all results to provide a single figure for reporting is to lose detailed information. When scores must be aggregated for reporting then we need to use models which result in the least loss of information and to make the rules explicit.

To summarise, I feel the program should consider the learner as an active participant, use a range of formative assessments, with a focus on self-assessments, feedback and leaners’ progress rather than awarding grades.

What are the benefits, issues of using self-assessments and are they reliable?

To evaluate self-assessments, the literature was reviewed to establish the associated benefits and issues, in addition to the reliability of the tool for sour situation.

Several benefits of using self-assessment have been identified. A number of language researchers have found self-assessment to be a reliable method of improving students’ language skills and abilities (Ekbatani, 2000; Nunan, 1988), developing learner autonomy and metacognitive engagement (Andrade and Du 2007; Cassidy 2007), enhancing learning, including deep and lifelong learning (Taras 2008) and it contributes to student achievement (Hughes, Sullivan & Mosley 1985; Schunk, 1996; Ross 2006). Studies have also shown that self-assessments has a positive effect on students’ learning motivation (Pope, 2001) and learning performance (McDonald & Boud, 2003).

However, several issues with self-assessment have also been identified. Some students are reluctant to self-assess, feeling they lack the necessary skills, confidence or ability to judge their own work or simply are afraid of being wrong (Leach 2012), preferring and expecting to be assessed by experts (Evans, McKenna, and Oliver 2005) or students may see it as the teachers’ responsibility (Brown and Knight 1994). In addition, in many Asian countries the very concept of self-assessment goes against deep-rooted cultural expectations about learning and giving themselves a good grade is considered inappropriate, boasting (Leach 2012), resulting in individuals from Eastern cultures generally displaying a modesty bias, and thereby underrating their performance (Yik, Bond, and Paulhus, 1998). Therefore, it is important to explain the rationale to the learners and demonstrate that as learners we daily self-evaluate (e.g. reciting a list of words). To address cultural issues it may require individual consultations to allay concerns.

Although self-assessment is being used in a range of settings: science, maths, and language classes; primary, secondary and tertiary education; there is still some doubt about its reliability which Gipps defines as ‘the extent to which an assessment would produce the same, or similar, score on two occasions or if given by two assessors’ (1994: p. vii). Bachman and Palmer (1989) found that a group of EFL learners in the US were able to reliably self-rate themselves for their communicative language abilities. Boud and Falchikov (1989) found there was no consistent tendency to over or underestimate performance by students. Some students in some circumstances tended towards one di­rection, others in the same or different situations towards the other. However, they found the ability of self-assessors was a noticeable variable, with the more able students making more accurate self-assessments than their less able peers. Weaker and less mature students also tended to overrate themselves and the weaker they are, in terms of teacher ratings, the greater the degree of overrating. One explanation offered by Boud and Falchikov (1989) for this was learners not being aware of, or choosing not to subscribe to, the standards set by teachers, erred on the side of optimism. Boud and Falchikov (1989) also found that over-estimates are more likely to be found if the self-assessments contribute to the student’s grade in a course and young children may over-estimate due to a lack of cognitive skills to integrate information about their abilities and are more vulnerable to wishful thinking. Ross, et. al (1999) found that agreement of teacher and student assessments is higher when teachers provide direct instruction to students on how to self-assess their work, Ross (2006) says that the strengths of self-assessment can be enhanced and weaknesses addressed through training students how to assess their work thereby placing training as central to the successful implementation of self-assessment. According to Ross (2006), one other factor which may be overlooked by teachers is that students may include in their self-assessments information that is not available or obvious to the teacher, such as effort. We have found that discussing with the students their grade helps to draw out underlying beliefs of the students on their work, rather than relying totally on the physical evidence presented.

Issues identified with previous studies

Ross (2006) and Boud and Falchikov (1989) after extensive review, both found a lack of sufficient studies looking at improvement over time, to draw any firm conclusions and there is particularly a lack of studies on the influence of practice on self-marking over time. They also expressed some concern about the quality, especially regarding the lack of definition in the criteria used by teachers and students, something we address later in the essay.

What needs to be consider before Implementing Self-Assessment

Considering what the literature provided, outlined below is the approach taken on implementing self-assessment in EPP. Firstly, as identified by Boud (1995, p.189), ‘an effective program needs to gain student commitment, link well with the subject matter, and encourage students to take greater responsibility for learning’. A common issue on many language courses which follow the objectives approach, is only the products of learning are assessed which is insufficient to guide learning. However, on the EPP the process of learning is often of greater importance than specifically what is learned as not all learning is evident in a final product, no matter how well thought out the assessment tool may be.

We will look at now how negotiating criteria and the selection of evidence can involve learners more in the assessment process.

Negotiating the Criteria

Boud (1986), maintains that the involvement of learners in making decisions about the criteria which are appropriately applied to their work and their making of judgements about achievements is the key characteristic of self-assessment. He further says that engagement in such activities helps to stimulate metacognitive skills and wean students from dependence on the assessments of others. Boud (1995) suggests two approaches to generating criteria; structured group activities and structured written schedules. The former is used to generate common criteria for a class, and I have found it a good way to start the process as the class generates and discusses potential criteria for inclusion. This can address some of the issues identified earlier such as; students’ reluctance to self-assess, supporting less able students, and poorly defined criteria. In general, for writing tasks there are four areas that are looked at, Grammar, Lexis, Coherency and Content. The discussion can help students to become more aware of the criteria and assist them in the structured written schedules, which consists of three steps to guide each student in individualising the process.

These steps are as follows:

• identifying the criteria which they consider appropriate to apply to their work, for example they may choose a number of the four areas or another such as format, thereby taking responsibility for learning and personalising it;
• clarifying these criteria, what specific area for example are they examining; and
• assessing the priority or emphasis which they wish to give to each criteria, encouraging deeper thinking and learning

Once satisfactory criteria have been generated, students then use them to judge their own performance. From a checklist of the criteria, students may simply award themselves a mark with respect to each criterion chosen, and then make a statement justifying that mark (e.g. on format, I have written four paragraphs, including a clear introduction and conclusion). The emphasis on which criteria is important to the learner will change over time as they identify additional areas of weakness or choose to challenge themselves.

The initial negotiation of the criteria occurs within the first two weeks of the program so learners can become familiar and start using them as quickly as possible. Learners discuss with teachers the criteria that they have chosen to be used in assessments, to eliminate potential confusion and to avail of support.

The selection of evidence.

The learners are involved in deciding on the form and quantity of evidence to be used in assessment, which allows for individuals to take more responsibility, by selecting from their own work and creating a portfolio, which will be explained in greater detail later. Learners indicate what evidence they have chosen, for the attainment of their goals, including essays written, as well as feedback received, and reflections.

The negotiation of learning goals.

Historically, the majority of assessment tools have been created based on external goals and imposed on the learners usually by the curriculum. However, it is important and appropriate that students are actively involved in setting class or personal goals and assessing themselves so that through this, they are more invested in learning and develop the skills required in how to learn, leading to the development of independent learners and critical thinkers. Self-assessment can provide a very personal and detailed record of learning. This negotiating of goals occurs early in the program so that the purposes and directions of the program are set to meet the learners’ self-perceived needs. Initially these are quite general but over time become more refined and individualistic as they are reviewed or change during the course. Goals may relate to the process as well as the outcomes of the course. Goals should be of a personal or context-specific nature (e.g. I want to be able to write a personal statement).

Assessment approach on EPP

On the EPP we have endeavoured to compile an assessment for learning approach, which supports the teaching/learning process, rather than assessments of learning which, simply measures student performance by tests and examinations. Tasks, which support higher order skills and support learners’ learning goals are utilised. Examples of tasks used are as follows; written essays, role plays, maintaining vocabulary and reflective learning journals. Some of the characteristics, and underlying reasons for the tasks are as follows:

• a clear rationale for the activity, so that learners can be actively engaged with a task which they accept is for learning (not passively following a set of instructions);
• explicit procedures so learners know what is expected of them, both in carrying out the tasks and in self-assessing; given that these were mainly new experiences for them, and that lack of training was identified as a major failing in previous studies;
• the task is constructed to allow significant elements of choice by the learners so that they can begin to own it and make it meaningful and worthwhile for them, taking greater responsibility for their own learning and learn to become independent of their teachers (Boud 1988);
• selection and reflection elements which reinforce student responsibility in taking charge of their learning and it provides a more valid, individualised assessment (Boud, 1995);
• reassurance so that learners can be honest about their own performance without the fear that they will expose information which can be used against them, and to address and cultural sensitivities or bias;
• tasks contribute to the final grade, although, based on the findings of previous studies, there is the potential for learners to overestimate or underestimating due to cultural bias, it was deemed necessary to do this to because of the general need for recognition by learners and to comply with external bodies;
• allocation of class time to complete the tasks and to enable learners to get assistance, this alleviates time pressures on the leaners and to also allow for sufficient training/retraining to take place so learners gain confidence in the process;

Finally there was an emphasis on the process rather than just the product of assessment (Boud 1995). Students on the program are allowed to work and re-work their written drafts, based upon multiple sources of feedback, until such time that it is ready to be submitted allowing for student’s writing ability to be assessed in an ongoing, authentic context.

Teachers were advised to exercise caution as tasks are usually completed over a period of time, both inside and outside the class and the degree students are assisted with feedback has the potential to affect validity (Boud, 1995). To address reliability, moderation where teacher’s and student’s scores are compared, can be used. However, if teachers moderate student’s results excessively, then students do not put much effort into being objective but simply rely on the teacher to do the assessing. At the same time, if teachers place the full responsibility on students, the danger is that there will always be some students whose self-assessment is not justified, however, the payoff is that the majority of students undertake their self-assessment much more seriously, and therefore learn a great deal more in the process of doing it (Boud, 1995).

Portfolios of Evidence

A recent trend in language assessment advocated by Boud (1995) and Race (2001) is the inclusion of portfolios in a course. A portfolio allows students to track their progress by compiling a selection of their work, selected from larger body of work. The portfolio is then presented with feedback comments and a reflective piece written by the student to justify the selection. Race (2001) points out that while most other forms of assessment are like ‘snapshots’ of particular levels of development, portfolios can illustrate progression, reflecting how quickly students can learn from and implement feedback. The involvement of the student in reviewing and selecting is central, helping ‘student-centred learning’ to become a reality (Calfee & Freedman, 1996). Kathpalia and Heah (2008), stress the importance of reflection stating that “a writing portfolio without reflection is merely a collection of written work which does not contribute to real learning”.

Portfolio Procedure

At the beginning of the program, time is taken to ensure learners are provided with instructions on how to reflect on activities as well as filling self-assessment checklists through which they could improve their autonomy in writing. As identified earlier, training is key to the successful implementation of self-assessment. After the first two weeks, a simple class checklist was created for the purposes of self-assessment. An exemplar piece of writing was then given to the individual learners and they were asked to use the self-assessment checklist with this writing. The results were discussed with the class along with individual consultations. This process was then repeated with another piece of writing. Once learners were familiar with using the checklists, they could create their own, using the individual criteria chosen by themselves. After that, learners were then given a list of topics to choose from and were required to write one task during class and one outside the classroom. Learners then had to assess their work using their individual checklists. Again, the general results were discussed with the whole class in addition to individual feedback sessions. After one month learners showed significant improvement in self-assessing as confidence grew and could begin to self-evaluate their own work independently, using their own checklists, and to add them to their body of work.

Grades were awarded by the learner based on their individual, agreed negotiated criteria. Initially, each student awarded themselves a grade together with a justification for it based upon the evidence submitted. Students are encouraged to consult with their peers if they are lacking confidence and to focus on the process of evaluation rather than simply the grade. Independently of this, a teacher assessment is made using the agreed criteria on the evidence available but without knowledge of the student’s proposed grade. If the two proposed grades did not fall within the same band, a discussion took place during which each party justifies their grade. Agreement generally resulted, but there is provision for final arbitration by a third party if needed.

Race (2001) argues that portfolios can be high on validity as it is possible to assess appropriate evidence of achievement relating more directly to intended learning outcomes, than (for example) can be achieved just with written exams. Race continues by saying that portfolios contain evidence reflecting a wide range of skills and attributes, and can reflect students’ work at its best, rather than just a cross-section on a particular occasion, such as one-off exams. He cautions that ownership of the work can sometimes be in doubt and the inclusion of an oral assessment or interview, can validate the origin of the contents of portfolios. There are however, some issues with portfolio use, with McMillan (2004) and Race (2001) highlighting that assessing portfolios is time-consuming, requiring time for both designing the portfolio and preparing rubrics for scoring. In addition to that, the teacher has to train learners to self-assess their work adequately, which often entails a one-to-one conference with each student so that portfolio implementation is done properly. As McMillan puts it, portfolio assessment ‘requires time, expertise, and commitment’ (2004, p. 238), all of which are not always available.

Reflective Learning Journals

Reflective learning journals are another important aspect of self-assessment and the EPP, with evidence showing that good learners have better metacognitive processes than poor learners (Ertmer and Newby, 1996). Developing reflective skills is an important aspect of self-assessment, leading to a ‘reflective practitioner’ which according to Schön (1987), includes: reflection-in-action that is immediate, short term, concerned with a devising a new strategy for approaching the situation; and reflection-on-action, typically undertaken some time after an event has occurred. The challenge we had was ways of incorporating reflective activities in the course. Boud et al, (1983) suggests that learners maintain a journal, to reflect on their learning, over a sustained period, maintained with the intention of improving or supporting learning. Records can include both academic as well as personal development with students generating records on such items as: their objectives and how these have been addressed and achieved; expectations, attitudes, values, beliefs, and skills. The records can start off structured but may become more unstructured as learners take ownership. Morrison (1996) identified some matters which need to be considered regarding the reflective activity: not all students find reflection easy; there may also be cultural issues where the concept is particularly difficult to grasp; and what is the depth of reflection required. A means of addressing these is to provide real examples of reflective writing, as well as some structured questions to the learners. Allowing for adequate practice and providing opportunities for feedback can also alleviate any potential problems. In terms of assessing the work, initially a journal may be considered ‘satisfactory’ and passed, or ‘not yet satisfactory’ and not yet passed, avoiding some of the difficult judgements about work that may be very diverse and / or creative (Morrison 1996). We found that this can also encourage greater participation as leaners feel that they are not being scrutinised about what they are writing.

Evaluation of using Self-assessment in EPP

One of the greatest challenges was with the concept of criteria where both the teachers and learners preferred to rely on well-known externally imposed criteria rather than take ownership of self-generated criteria, negotiated in the classroom. Indeed, there was also a strong resistance from the institute itself, however the future success of this approach the acceptance and willingness of all the participants.

From an learner’s perspective

Each term, a survey concerning the course in general and the use of the self-assessments is conducted to gather the views of the learners. For the most part, after the initial introductory phase, learners find self-assessment a useful tool which helps them focus on their own learning:

I found it very difficult in the beginning but now I know it will be good for my future study.

Difficult to start but then I started to enjoy it when I realised what I had achieved.

Students commented on the tasks in general (reflective journal; self-assessment):

A very interesting and different experience for me.

I learned how disorganised I am and that I need to change.

I now have more confidence in my work before I submit it.

Students were very positive on the experience:

Amazing, I had no idea that I had achieved so much until I reviewed my journal at the end.

I now really enjoy reflecting on what I have done not just memorising information.

Students rarely, if ever find the task of self-assessment easy, especially in a Chinese society where the teacher is traditionally viewed as having ultimate control. Some learners are naturally more self-reflective or self-critical than others, and some are more willing to share their learning than others. It provides an opportunity for students to reflect on their learning and think about the applications of ideas in their own situations. It is common for them to report that they only start to become aware of what they have learned when they looked back on the course in a systematic fashion, in preparing to submit their portfolios. Two major obstacles that learners had difficulty overcoming was the doubt in their ability to assess themselves and the objection to the concept of self-grading, arguing that grading should be the sole responsibility of the teacher, which is similar to studies mentioned earlier (Leach 2012, Brown and Knight 1994) however, we feel this has been addressed.

From the Teacher’s Perspective

Despite the increase in using self-assessment, Ross (2006) explains that teachers still retain doubts about the value and accuracy of the technique, saying many teachers holding the view that learners are incapable of self-assessment believing that learners are unable to appreciate or understand the process. In our situation some of the more senior teachers resisted the change in the power dynamic that self-assessment entails as it not only changes the role of the teacher but also the relationship between the teacher and learner. A secondary issue we experienced was when the teachers themselves are unsure of or are having difficulty in interpreting criteria and are therefore reluctant to negotiate with the learners. However, I have found that through the process of discussion the criteria ultimately become clearer.

From my perspective

Producing a portfolio of evidence has the advantage of students summarised and demonstrating their learning at many different stages of learning and has been a valuable takeaway from the program for the learners. Now before submitting a written piece of work, many learners have gone through the process of self-assessing and therefore have formed an educated opinion of how good they think the work is which leads to reduced anxiety. Overall, assessment portfolios are beneficial to students. They give them the opportunity to reflect, and to develop their abilities in assessing their own work and understanding. Thus, learners end up eventually taking responsibility for their own learning and have continuing opportunities for using their creativity and imagination and increasing the quality of their work (Barton and Collins 1997). On reflection, I believe that although portfolios require considerable work on the part of both the students and the teacher, they provide a much more effective assessment tool than those used traditionally because the ongoing and developing nature of the portfolio provides a much clearer indication not only of what the learners have achieved (the learning process) but also what the teacher has enabled the learners to achieve (the teaching process and product).

Conclusion

The objective of this paper was to evaluate using self-assessment methods in the EPP program.

Although self-assessment has been largely ignored in pedagogy until recently, we have attempted to move away from the objectives/ behaviourism model and over many cycles of modification in collaboration with different teachers and various classes of students we have developed an approach to self-assessment which seems especially suited to courses which emphasise student autonomy and reflective learning. Although the research and our experience of self-assessment does pose some concerns regarding reliability and validity, discussion and training over rubric construction, and grading can go some way towards alleviating this. Despite some concerns there is strong evidence (Hughes, Sullivan & Mosley 1985; Schunk, 1996; McDonald & Boud, 2003; Ross 2006) that self-assessments yield consistent and positive results and we are reasonably confident about its benefit to our program as an assessment tool. Ross (2006) sums up our sentiments towards self-assessment by saying that “teachers who make a serious commitment to learning about self-assessment and teaching these techniques to their students can plausibly anticipate enhanced student motivation, confidence, and achievement”.

Going forwards, as the range of assessments used increases on a course, professional development programmes which develop teachers’ skills in observation, questioning and interaction are needed so these approaches can be integrated into regular classroom teaching and assessment. The obvious potential of self-assessment for language institutions and classrooms as both a valid and reliable supplement to traditional assessment is clear and will likely increase as more courses adopt them.