The Four Purposes of Assessment: Competitive, Values-signalling, Supportive and Ipsative

By Terence Tay Jia Ming, SSD, Lakeside Primary School 

Assessment can serve many purposes. It is helpful to keep these broad purposes in mind as we continue to improve our assessment practices.

 

Competitive achievement

Traditionally, assessment (of learning) has served as a sorting mechanism, seen as an objective and fair way to allocate students to different classes or different schools. Within the milieu of academic meritocracy in Singapore, good academic grades have been seen as the means to obtaining more resources and a better quality of life through better-paying jobs. From society’s point of view, as exhorted by then Prime Minister Lee Kuan Yew to Principals of Schools in Singapore (1966), it was also necessary to identify top leadership to be nurtured in elite schools, the middle strata of good executives and the broad masses. There is considerable prestige to be gained from doing well in one’s studies. High achievers have often been recognized through awards, celebrating their successes. Yet, because there is a finite number of places in sought-after schools and the number of awards that can be given, it becomes a zero-sum game due to the bottleneck; only a few can ‘win’. Assessment becomes competitive due to norm-referencing, and it becomes high-stakes and therefore highly stressful.

 

Competitive assessment has its role in our system. Being able to identify elite achievers can help us identify leaders who can lead excellence in different fields, which is important for us as a nation to thrive. A healthy dose of peer competition, especially for elite achievers, can also bring out the best in people, raising standards.

 

However, it should not be the dominant purpose of assessment for the broad masses. In this frame, over time, assessment becomes the goal. Students partake in learning activities to do well on the tests. Teachers, in caring for students, get better at teaching to the test and helping students to score well. Learning becomes less enjoyable for many, especially in the lead-up towards national exams, with effort inflation (e.g. increased load of practice papers and possibly tuition lessons). Also, if (academic) assessment outcomes determine the personal worth in society, then it can have negative consequences on students’ developing identities. “I’m not good at English. I’m not good at learning. I’m a failure.” These are not sentences we as educators hope to hear from our students. Thankfully, as a system, we have been shifting away from an emphasis on this purpose.

 

Values-signalling

Assessments can signal what we value. For example, by recognizing success in non-academic domains such as in leadership, service, CCAs and character, we show what is important for us as an education system; we broaden the definition of success. Within assessment tasks, we can also signal what is of value. In national exams and then cascaded down to school assessments, we see the moving away from a regurgitation of facts and procedural knowledge towards a thoughtful application of learnt concepts. As one of the lecturers, Dr Tay Hui Yong, quipped if teachers are going to teach to the test anyways, let us make sure that the tests assess what we find valuable. For example, in the mathematics syllabus,

communication is an important process used to express mathematical ideas. Assessment tasks should therefore provide opportunities for students to communicate, and to be assessed on their communication.

 

Supporting achievement

In order to signal what we value through assessment, we must actually be clear about what we value. Professor Kelvin Tan has proposed a triangulated model of Assessment for Learning (2013), which has three components.

 

The vertical line refers to criteria and standards. Students (and teachers) need to have clarity about where they are going with their learning. Which constructs are important to develop for each subject discipline? Which constructs are important to help the school achieve its vision? Beyond curriculum content standards, what enduring understandings do we want to impart to our students from the total school learning experience?

 

In clarifying constructs, one thing we can do is to clarify standards of achievement. Sadler (1987) suggests four methods: (1) numerical cut-offs, (2) relying on assessors’ tacit knowledge, (3) use of exemplars, (4) explicit descriptions (with different levels either differentiated by number of features, or by distinctively different features). Explicit descriptions in combination with exemplars seem to work best.

 

Reflecting on my own teaching practices, the planning process usually goes from specific instructional objectives to learning activities to the setting of assignments in order to check for understanding and provide feedback. Perhaps that sequence is in part reinforced by how the Singapore Teaching Practice organises the four teaching processes, with the teaching areas under ‘Assessment and Feedback’ coming in at the end of the list. Perhaps the Curriculum triangle of Content-Pedagogy-Assessment has also contributed to the impression, with ‘Assessment’ being mentioned third. Shifting the paradigm, we should instead consider front-ending assessment. From specific learning objectives, we identify what evidence of successful learning looks like before we design learning activities that will elicit successful learning.

 

Ipsative achievement

Assessment should help students learn. It should help students grow from each learning experience and be better than they were before in whatever the focus of learning is. Apart from knowing where they are going, students also need to know how they are going and where to next (Hattie & Timperley, 2007).

This would therefore depend on the effectiveness of the assessment task design to gather assessment information and the effectiveness of assessment feedback practices to allow information to be used to improve learning – the other 2 sides of the triangulated model mentioned above.

 

For assessment task designs to promote the intended learning, we need to bear in mind assessment validity, so that the inferences we make are accurate and the subsequent actions we take are defensible. We look out for construct validity. In assessing for the knowledge, skills or attributes through the tasks, we are mindful of construct underrepresentation and construct-irrelevant variance.

–    Construct underrepresentation: Have we included all important dimensions or facts of the construct? For example, in getting students to solve Mathematics problems, we may focus on their solution, but we may neglect the way they present the solution, which is part of Mathematics communication, an important component of Mathematical problem-solving. To assess a wider range of educational outcomes, we probably will have to design alternative assessments such as performance tasks, presentations and projects (beyond pen and paper assignments)

–   Construct-irrelevant variance: Have we added irrelevant dimensions? For example, we may award or deduct marks based on the timeliness of submission or neatness of handwriting, or we may include a component related to the aesthetics value, which may not be part of the construct.

 

For assessment feedback practices to be useful, we need to bear in mind the kind of feedback we are giving. Hattie & Timperley (2007) suggests that feedback can be directed at four levels:

–        Self: Help students feel good or bad about themselves; unrelated to the task

–        Task: Help students improve on the current task

–         Process: Help students improve performance on subsequent related tasks, by improving the processing of information

–      Self-regulation: Help students improve their self-efficacy, self-regulatory proficiencies and self-beliefs about themselves  as learners

 

The aim is to move towards providing feedback at process level and self-regulation level. Process level feedback can improve disciplinary thinking, while self-regulation level feedback can help develop students as self-directed learners, which is a desired outcome of education.

 

Improving feedback uptake

For assessment to have an impact on student learning, we need to concern ourselves with feedback uptake. No amount of quality feedback will be useful if students do not act on them. Another paradigm shift we, therefore, need to make is to see feedback not as information but as sense-making (Winstone & Carless, 2019). It is not just about what information teachers gather about students’ learning, how they do it and what they do with the information. It is about giving serious consideration to students’ recipience of feedback. Students need to be actively involved in the whole assessment process (feeding up, feeding back, feeding forward).

 

Feed-up: To understand success criteria and recognise quality work, students should be involved in analysing exemplars, which can be from teachers’ modelling, peers’ work, or past students’ work. They can have a hand in co-constructing the success criteria and providing peer feedback based on the criteria.

 

Feedback: Students should be given opportunities to make sense of feedback and seek additional feedback. Time for processing feedback can be provided in class, where the teacher and peers are available to be resource personnel. If class time is limited, we can tap into technological affordances, such as discussion forums, where students can provide each other with additional feedback.

 

Feedforward: Students should be given opportunities to respond to feedback. This can happen through task design. Some task designs that enable uptake include two-part tasks, patchwork design, task series (i.e. a series of similar tasks), and draft-plus-rework (Winstone & Carless, 2019).

 

All in all, there needs to be dedicated time for student dialogue, reflection, and improvement. To engender uptake, teachers not only have to concern themselves with assessment design to involve students. Because the giving and receiving of feedback can be a vulnerable and emotional experience, teachers have to look into relational sensitivities. There needs to be a psychologically safe environment created, where there is mutual respect and the focus is on improving learning and not on judging. Teachers can tap on a feedback sandwich model (e.g. praise, question, polish) to soften the impact. Teachers need to develop teacher feedback literacy (Carless & Winstone, 2020).

 

At the same time, we can help students develop feedback literacy so as to enhance not just active, but proactive recipience. According to Carless & Boud (2018), there are four components to student feedback literacy.

(1) We help them appreciate feedback – to recognise the critical role that feedback plays in their own learning.

(2) We help them manage affect – to recognise that there is an emotional component to giving and receiving feedback, 

     which needs to be managed.

(3) We improve their ability to make judgments by involving them in the whole assessment process, as mentioned 

     above.

(4) Finally, we help them see that they need to take action for feedback to work.

 

When students develop feedback literacy, they not only improve in their learning, but they become metacognitively aware that they are learning. They learn how to learn, which sets them up for lifelong learning.

 

Leading assessment changes

Effective professional development

How might we induct teachers into these new ways of thinking and doing? Compared to researchers and teacher educators at NIE, teachers are in the privileged position of having easy access to students. They can test out ideas and subsequently refine or reject them. They can do so in a community, collaboratively dialoguing about ideas and developing resources. This need not be at a whole-school level. When we return to school, our own classrooms can be testbeds from which we can gather some evidence of success or failure. These can be stories we share with colleagues to invite participation in this journey of assessment literacy. What touchpoints do we have available to discuss assessment practices? Apart from formal professional development sessions, our subject-based sharing sessions and assignment conversation platforms can be tapped on.

 

Anticipating resistance

One thing to recognise is that teachers back in school do not have the same luxury of time and space as we have had to learn and to critically reflect. We go back with eyes wide open, anticipating different kinds of resistance. During MLS, we have learnt about four archetypes of resistances, which we can plan our responses for.

Unhelpful trait	Characteristics	Leadership Responses
Helpless	Only cares about exam results; cites high-stake national exams to prepare students for	Highlighting dual accountability of teachers: not only to help students do well for exams, but also to help students have a holistic education.
Defensive	Blames students; does not believe students will be motivated enough to be involved	Work with Senior Teachers teaching low-progress students to see how to tweak assessment designs and feedback to suit students’ needs
Too Busy	No time for any change	Teachers need to see exemplars (for a concrete understanding of intended change) and evidence of success (e.g. impact on students) Highlight problems of current practice
Superior	Knows it all; been there done that	Assess the validity of the teacher’s claims. Invite the teacher to be involved in developing others.

 

Shaping culture

Another source of resistance can be the school’s or department’s culture if there are mixed messages being sent about what we value. To lead assessment changes, we need to send clear signals about the value of changing assessment practices. The signals can take the form of explicit communications during professional development sessions, or they can take the form of assessment-related processes. For example, book checks are a common tool used by key personnel for quality assurance. However, if the quality is defined as a fixed amount of work to be completed, or perfect corrections by all students, then practices will remain status quo. More importantly than being accountable to the organization and to parents, we want teachers to be accountable to the learners. Must all work be completed and marked, and all corrections be cleared? This is a dialogue to be had in school. Otherwise, trying to introduce something like the 4 levels of feedback will just mean more time spent marking, which teachers will resist.

 

Conclusion – exercising technical, tactical and ethical leadership

According to Dr Tay Hui Yong, Assessment Leadership is the application of assessment literacy to inform educational administration, decision-making and change management. As Assessment Leaders, we exercise three kinds of leadership: technical, tactical and ethical (Tay et al, 2020).

 

In exercising technical leadership, we ensure the quality of assessments. We consider criteria such as validity, reliability and inclusivity (i.e. whether our assessment practices are free from bias and fair for all). We see if our assessments pass Socrates’ triple filter tests of truthfulness (i.e. true to purpose), goodness (i.e. do good; promotes well-being and motivation) and usefulness (i.e. meaningful; actionable). We stay updated with developments in assessment (e.g. through AFAL bulletins), introduce teachers to updated assessment literacy concepts and encourage the application of learning.

 

In exercising tactical leadership, we strategise the best way to achieve assessment changes, such as being grounded on research, tapping on available resources such as PD platforms & social capital, and managing resistance. We find ways to make it easier for the uptake of new ideas – for example, to help teachers manage practicalities such as the lack of class time, by tapping on technological tools for teacher and peer feedback.

 

Finally, in exercising ethical leadership, we focus on the moral imperatives of shifting paradigms. We look out for consequential validity of our assessments – intended positive outcomes, intended negative outcomes, unintended positive outcomes and especially unintended negative outcomes. We persist in the change initiative even if there is no immediate visible improvement in existing measures of student outcomes (e.g. performance in weight assessments and national exams) because not everything that matters can be measured.

 

Assessment is integral to the teaching and learning process. Improving assessment practices holds promise to maximise the return on learning for students, and to be commensurate with the investment of teachers’ time and effort – it is definitely a worthwhile endeavour.

 

References:

Carless, D., & Boud, D. (2018). The development of student feedback literacy: enabling uptake of feedback. Assessment & Evaluation in Higher Education, 43(8), 1315-1325.

 

Carless, D., & Winstone , N. (2020). Teacher feedback literacy and its interplay with student feedback literacy. Teaching in Higher Education, Advanced online publication

 

Hattie, J., & Timperley, H. (2007). The power of feedback. Review of educational research, 77(1), 81-112.

 

Sadler, D. R. (1987). Specifying and promulgating achievement standards. Oxford review of education, 13(2), 191-209.

 

Tay, H. Y., Tan, K. H. K., Deneen, C. C., Leong, W. S., Fulmer, G. W., & Brown, G. T. (2020). Middle leaders’ perceptions and actions on assessment: the technical, tactical and ethical. School Leadership & Management, 40(1), 45-63. Tan triangulated model

Winstone , N., & Carless, D. (2019). Designing Effective Feedback Processes in Higher Education: A Learning Focused Approach.

Routledge.