Formative assessment

The objective of FaSMEd’s Work Package 2 is to establish a baseline of data on the approaches to low achievers in maths and science in the EU and South Africa. The Work Package leaders have asked the questions below (in bold) and some responses have been added from published policy and research documents. Note that this is work in progress.

Please comment below (you don’t have to register or leave your email address).

What is the general view of formative assessment in South Africa?

ANA report
It seems that formative assessment is recognised as important in the official rhetoric. For example, the Annual National Assessments (ANAs) aim to be used in formative ways. The 2013 Annual Report on these states that:

“The evidence in this report must be built into normal teaching programmes and also be used to inform specific interventions to improve the levels and quality of learner performance in schools. The report also provides SMTs with objective evidence to identify areas where individual teachers need specific support in terms of both content knowledge and various methods of facilitating learning. The identified areas of required support will also influence the choice of relevant teaching and learning support materials.” (p.7)

However, research (e.g. Mkhwanazi 2011, Khosi, 2009) has shown that teacher knowledge of what formative assessment is, and how they can use it in their practice, is variable. Many teachers do not understand how to use assessment to inform their planning and do not provide constructive feedback for their students. Khosi further points out that even when teachers know how to use assessment in formative ways, large class sizes mitigate against its effective use.

Have there been any local, regional or National studies or initiatives in South Africa concerning formative assessment?

The same report (above) provides suggestions for improving teaching and learning in mathematics in two sections. The first is “Effective teaching and learning” and the second is “Using assessment to inform teaching”. In the section on using assessment, it states:

  • Analyse learners’ errors and/or misconceptions to understand their thinking processes. The errors learners make may be as a result of a misunderstanding of what the teacher taught or ineffective teaching methodology used by a teacher. The errors that learners make provide teachers with rich information that they can use to improve their teaching methodologies. Attending to these errors immediately when they occur will prevent them from developing into misconceptions. Effective assessment does not only focus on the correct answer (product), but on Mathematics procedures (processes) too.

  • Invest in formative assessment (assessment for learning) because it paves the way for enhanced learner performance when assessed using summative assessment.

  • Align assessment to teaching i.e. learners should be assessed on what they have learnt and how they were taught. In other words it will be highly unlikely that learners will perform well when asked to conduct an investigation if they were never taught through an investigative approach.

  • Expose learners to various types of questions such as multiple choice questions.

  • Expose learners to questions that are pitched at different cognitive levels such as questions that require straight recall (knowledge), routine procedures, complex procedures and problem solving. The main aim is to narrow the gap between formal or informal assessment that is internally set and conducted by teachers in different schools and external assessment such as common examinations that are set at provincial level or the Annual National Assessment set at national level.

5 thoughts on “Formative assessment”

  1. Please see CAPS doc per phase (select grades 7 – 9 for fasmed). At the end is section on assessment which is helpful and % given to include in formal assessment on each thinking level. The new requirement has increased to 45% of the higher thinking levels. Most teachers in our country is not used to this.

  2. The SA official National Curriculum Statement Curriculum & Assessment Policy document (CAPS) requires all teachers to adhere to pacesetters that specify hour by hour what is to be taught and when. The required time for mathematics is 180 hours per year of which 30 hours must be given to formal assessment. It is normal in government schools for all lessons for all grades to be suspended during two examination periods in the year and children may not attend school unless they are writing an examination. There is often little or no feedback to learners on these examinations.

    CAPS for Mathematics Grades 7–9 (p157) sets out 4 cognitive levels to be assessed: knowledge 25%, routine procedures 45%, complex procedures 20% and problem solving 10% . School based assessment (tests, examinations, assignments, investigations, projects) makes up 40% of the formal assessment and the end of year public examination 60%.

    (Note: ‘promotion purposes’ refers to decisions as to whether a learner should move up to the next grade at the end of the year. Many learners are required to repeat a year and this results in most classes having a wide age-range. In government schools all mathematics teaching is mixed ability up to and including Grade 9. All Grade 12 learners are required to be entered for the National School Certificate (commonly known as ‘matric’ ) in Mathematics (one level for all) or in Mathematical Literacy, so learners are split into these two cohorts in grades 10-12.)

    The SA policy on formative assessment is quoted below from CAPS Mathematics Grades 7-9 p.154-156. How far this is carried out in practice is another matter that I shall comment on later.

    “Formative assessment is used to aid the teaching and learning processes, hence assessment for learning. It is the most commonly used type of assessment because it can be used in different forms at and time during a Mathematics lesson, e.g. short class works during or at the end of each lesson , verbal questioning during the lesson. It is mainly informal and should not be used for promotion purposes. The fundamental distinguishing characteristic of formative assessment is constant feedback to learners, particularly with regard to learners’ learning processes. The information provided by formative assessment can also be used by teachers to inform their methods of teaching”

    “Assessment for learning has the purpose of continuously collecting information on learner performance that can be used to improve their learning.”

    “Informal assessment is a daily monitoring of learners’ progress. This is done through observations, discussions, practical demonstrations, learner-teacher conferences, informal classroom interactions etc. Informal assessment may be as simple as stopping during the lesson to observe learners or to discuss with learners how learning is progressing. Informal assessment should be used to provide feedback to learners and to inform planning for teaching, but need not be recorded. It should not be seen as separate from the learning activities taking place in the classroom.”

    “Self-assessment and peer-assessment actively allow learners to assess themselves. This is important as it allows learners to learn from, and reflect on, their own performance. The results of the informal daily assessment tasks are not formally recorded unless the teacher wishes to do so. The results of daily assessment tasks are not taken into account for promotion purposes.”

  3. In practice SA teachers, with the best will in the world, find it very difficult to use formative assessment to assist their learners because of time pressures, large classes and lack of understanding due to language difficulties.

    The CAPS pacesetters require teachers to move on from topic to topic week by week whether the learners understand the concepts or not.

    Classes of over 50 learners are very common and often much larger than that, sometimes over 100 in one class with one teacher. Often these big classes are crammed into classrooms built for 30-40 learners so learners have to climb over desks to get to their places and it is impossible for teachers to move around the classroom. All classes are mixed ability and there are almost no learning support assistants for special needs learners.

    There are 11 South African languages and all school examinations are written in either English or Afrikaans. It is usual for all lessons to be in the home-language of the learners in Reception and Grades 1-3 and then to be in English or Afrikaans from Grade 4 up. There is a significant difference between rural and urban schools because in the rural schools it is often the case that nobody speaks English outside the school classroom. See the many publications by Professor Jill Adler and colleagues on teaching mathematics in multilingual classrooms. However, unlike the western world, ‘multilingual classrooms’ in South African often means just two languages: mother tongue and official language of learning and teaching (LOLT), and frequent use of code switching. (I leave it to Lindiwe and others to add to this.)

    In hundreds of interactions with teachers from grades 4 to 12, from across the country, over 11 years, I have heard the same worries hundreds of times, namely: “learners seem to understand the mathematics during the lesson but then why do they do so badly in written tests?” So informal formative assessment is being used but teachers find it difficult to apply it to help their learners. In discussion it often transpires that teachers usually interpret questions in class to learners using code switching, and learners rarely have to read the question in the LOLT by themselves except in formal written tests.

    Quizzes for Grades 10-12 Mathematics, provided through Nokia sponsorship, for learners to complete on their cellphones, along with some explanations of the basic mathematical concepts and processes, have helped teachers with formative assessment. This has been effective where teachers have used the Mobile Maths functionality to register their class and then to use the automatic monitoring marksheets to check on the progress of their learners. (Nicky Roberts will be able to give more information here).

  4. The National Protocol For Assessment Grades R-12 sets guidelines for formal and informal assessments in the classroom. It aims at providing guidance for “an indication of learner achievement in the most effective and efficient manner”. Assessment for learning (AfL) is encouraged as a tool for increasing learners’ progress though emphasis is largely placed on the use of formal assessments to provide teachers with a “systematic way of evaluating how well learners are progressing”.
    Teachers may have used aspects of AfL in their day-to-day teaching but there is little evidence on the ground that the teaching and learning strategies are used explicitly in the classroom as tools for identifying where learners are at, where they need to be and the steps they need to take for progress to be made. Our experience with teachers at AIMSSEC is that little is known or understood about progression in learning with much emphasis placed on curriculum coverage and pace setting in the classroom. This emphasis may be necessary for different reasons as indicated by the National School Effectiveness Study viz. the average number of curriculum topic areas covered in the best learners’ workbooks ranges from 11.8% to 34.5% in the provinces.
    Furthermore, the ANA 2013 Diagnostic Report and 2014 Framework for Improvement gives a national diagnostic summary of the key content areas identified as areas of weakness. Learners’ responses are analysed for some questions with suggested interventions though lacking AfL strategies to facilitate improvement; the result being a continued traditional approach to teaching and learning. Although assessment is well documented, AfL is very much in its embryonic stage of achieving the desired improved results.

  5. A few things I think about when reading this:
    Firstly I think this would benefit from differentiating by Grade. I think there is much that is peculiar to the Foundation Phase about formative assessment, and which simply does not fit with how this is considered for older grades. So separating phases or at least recognizing the differences for each phase would be helpful. I reflect mostly on some of Foundation Phase issues I thinking about when looking at this.

    Secondly teachers need help with how to do formative assessments. There is much assumption in SA that formal written assessments give meaningful outcomes – I doubt this in the early grades where language is still in development, as is basic reading and writing (in the LOLT). In my experience individual interviews for early grade assessment give completely different outcomes to formal written assessments (like ANAs). So small group and one on one administration of assessment is a time consuming, burdensome but very necessary diagnostic process; and needs to be part of mainstream teaching time at Foundation Phase. This help is in understanding what is obtained when a child has to read and write for themselves, vs what is possible in the presence of a trusted and caring adult, and noting these differences. Big issues here are whether on not children are encouraged and or permitted to make use of manipulatives, rulers, empty number lines, bead strings, etc to support their calculations. Attention in assessments on the efficiency of strategies chosen, and what representations children use is important to the Foundation Phase teacher in planning how next steps for that child.

    The policy talks of different assessment approaches but in reality teachers assess through written tests and exams. What is a mathematics project? What is an investigation? How do they differ, how are they assessed? How do we do more meaningful problem solving work through these formats? I see data handling prohjects repeated from grade to grade shifting maybe from favourite colours to means of transport, to recycling projects without much increase in mathematical sophistication. What makes a good grade 3 maths project? What would be appropriate at grade 4?

    Thirdly teachers need help in varying tasks (assessment or otherwise). I see few teachers able to reflect on a task and either make it harder or easier. The tasks is taken as face value. it is seldom seen as exemplifying a class of examples, which fit into differentiated (and perhaps structured) example space. So given the chance to make something easier, or make it harder teachers are at times at sea, as to what could be changed. This has been a valuable activity in working with teachers (I think)

    Fourthly the help is also necessary for interpreting assessment results. Early grade teachers, in my experience, do not know how to do item analysis and see for example the percentage passing a question. The normal interpretation of results is on what the individual learner attained (and no consideration for what a particular item reveals about areas of weakness). Reporting on items of weakness is required by some provincial departments of education, but watching early grade teachers attempt this revealed that they needed support on how to analysis the data (basics of excel and considering an item at a time, rather than collated individual results). From this I have generic excel template to help analyse assessment data (used in WCED and Gauteng)

    The focus on formative assessment is important, but the current SA emphasis on standardised assessments (ANAs) creates opportunity on how this annual stress can be used constructively to inform teaching and learning. So decent analysis of ANA results (on an item by item basis) and to facilitate error analysis and CoP reflection on teaching when reviewing ANA scripts following marking, would be worth while. That ANAs are administered in the 3rd term, then becomes are strength where they are used as diagnostic / formative assessment tools for a grade/class in a school.

    Lastly I feel like we are in assessment over drive: “You cant fatten the pig by weighing it” I think they say in the UK? We seem to be constantly weighing and not doing enough teaching. I know the dichotomy (teach vs assess) is false when done well, but to many FP teachers assessment feels very burdensome and overkill on the eve of reporting and assessment needed (often this is not formative)…This is made worse at FP level where meaningful formative assessment necessitates individual engagement with each child (in a class of 45 or 50….), on each aspect of the CAPS (in FP which is ridiculously fragmented and itemised) and in each term… you get the problem.

    Which reminds me, teachers need help on how to weight assessment by content areas. I found that some teachers were setting individual tasks for each content area (as per CAPS) as they did not know how to record and adjust the weighting of questions to meet the CAPS requirement. This meant 15 FATS (Formal assessment tasks) per term as the requirement was for 3 FATS (but they had to separate out all 5 content areas into separate tasks). So 15 maths assessments in a 10 week term: go figure! When was there any teaching (never mind learning)?

    Then some random additions:
    Have you seen the GiZ report:
    Assessment of Learning Outcomes with Reference to Early Grade Numeracy
    in Socio-Economically Less Developed Countries
    Die Deutsche Gesellschaft
    für Internationale Zusammenarbeit
    Jeff Davis
    Yasmin Sitabkhan
    School-to-School International
    December 2012?
    May be useful to you?.

    Happy to give information about Nokia Mobile Mathematics too if this is helpful. This is at other end of the schooling system at Grade 10,11 and 12. And the context is quite distinct there.

Leave a Reply

Your email address will not be published.