Involving primary school students in the co-construction of formative assessment in support of writing

ABSTRACT

This paper describes the enlargement of the initial conception of formative assessment with reference to constructivist, sociocultural and situated theories of learning and the concept of co-regulation. It reviews research on student involvement in formative assessment practices (self-assessment, peer assessment, whole-class discussions of criteria and exemplars) in the area of writing, with a focus on primary school classrooms. Student participation in the co-construction of formative assessment is illustrated by qualitative observations from a study of a writing activity carried out in grades 5 and 6. The observations are discussed in relation to findings from other research and implications are presented for teacher professional development and for future studies of formative assessment of writing.

KEYWORDS:

• Formative assessment
• self-assessment
• peer assessment
• co-construction
• co-regulation
• writing
• primary school

View correction statement:

Correction

Formative assessment was initially conceived, in the Bloom (1968; Bloom et al., 1971) mastery learning model, as a procedure planned and implemented by teachers who define learning objectives, construct formative tests, interpret results in a criterion-referenced framework, provide feedback to students and propose appropriate types of remediation for any objectives not initially attained. Student participation, in this perspective, consists primarily in the execution of the proposed assessment tasks and the remediation activities. Questions have been raised for some time about the benefits of encouraging more active student involvement in formative assessments as a way of increasing learners’ cognitive engagement and motivation and thereby enhancing learning outcomes. Among the publications that initiated concern with this issue figure an article by Sadler (1989), who argued that instructional systems need to be designed to develop students’ self-monitoring while carrying out a learning activity, and the landmark research review of classroom assessment by Black and Wiliam (1998), which included a section dealing with students’ perception and use of feedback and with their participation in self- and peer assessment.

This paper concerns student involvement in formative assessment of L1 writing activities (i.e. production of texts in the language of instruction of the school system) in the context of primary school classrooms.¹ It highlights more specifically the question of student contributions to the construction of formative assessment processes and tools. The aim of the paper is, first, to review studies that have investigated this issue and, secondly, to present and discuss qualitative observations from a study in grades 5 and 6 illustrating how teachers can encourage student involvement and some of the consequences.

The paper has five sections. The first section summarises proposals for enlarging the initial conception of formative assessment. Section two reviews results of research on student involvement in formative assessment, with emphasis on the area of writing in primary school. The next section summarises key findings from a study in fifth- and sixth-grade classrooms of student participation in the co-construction of formative assessment processes and tools in support of writing, and then presents qualitative observations of classroom interactions and their outcomes. In the fourth section, these observations are discussed in relation to other research findings. The final section suggests implications for professional development activities for teachers and for future research on formative assessment of writing in primary school.

Enlarging the conception of formative assessment

Two developments have contributed to the enlargement of formative assessment as practiced by teachers and students in the classroom.² Both developments concern the role of formative assessment as an interface linking the functions of teaching and learning (Allal, 2020; Allal & Mottier Lopez, 2005). The first development is the reconceptualisation of formative assessment on the basis of other theories of learning than the neo-behaviourist perspective of mastery learning. Constructivist theories describe the cognitive and metacognitive processes that allow transformation of the learner’s knowledge and skills and that need to be supported by interactive practices of formative assessment (Allal, 1979). Sociocultural theories and theories of situated cognition provide frameworks for conceptualising learners’ active involvement in assessment dialogues, with teachers and with peers, and their participation in the assessment practices of a classroom learning community (Allal, 2007; Allal & Pelgrims Ducrey, 2000; Gipps, 1999; Hickey, 2016; James, 2008; Torrance & Pryor, 1998).

A second, related development concerns the concept of regulation as the defining feature of formative assessment. As proposed by Allal (1979; Allal & Mottier Lopez, 2005), formative assessment can involve three processes of regulation differentiated by their temporal contingencies: retroactive regulation (remediation in Bloom’s model), which means that students return to a task not mastered and carry out corrections or improvements; proactive regulation, which means that teachers and/or students use feedback from assessment to improve their approach to new instructional activities; interactive regulation, which means that students adjust their engagement in an on-going task through their interactions with the teacher, with peers and/or with instructional materials and assessment tools. Whatever the temporal continency of regulation, the question arises as to the relative importance of different sources of regulation. Some researchers, notably Andrade (2010), have placed student self-regulation at the centre of formative assessment. Models developed in the literature on self-regulated learning have been adopted as a basis for conceptualising the processes of active learner engagement to be fostered by formative assessment, and in particular by student self-assessment and peer assessment (Panadero et al., 2018). In order to describe how student self-regulation operates in conjunction with contextual sources of Allal (2007, 2020) has proposed reframing formative assessment as a process of co-regulation defined as the joint influence on student learning of sources of regulation in the learning environment (structure of the teaching/learning situation, teacher-student interactions, peer interactions, instructional and assessment tools) and the processes of student self-regulation. In this perspective based on concepts of situated cognition and learning, co-regulation is seen as an overarching concept that encompasses the interdependent relations between contextual sources of regulation and student self-regulation. This conception of co-regulation differs from that of other researchers (Andrade & Brookhart, 2020; Hadwin & Oshige, 2011) who consider co-regulation based on teacher interventions and student-teacher interactions as a support structure or a form of scaffolding that will allow students to become autonomous self-regulated learners. In a situated perspective, the aim of formative assessment in school settings is to develop students’ capacity to participate actively in increasingly complex and diversified forms of co-regulation: in other words, to allow students to become increasingly competent co-regulated learners rather than autonomous self-regulated learners. In the formulation proposed by Adie et al. (2018), student agency operates within temporal-relational contexts of assessment activity but also has the potential to transform assessment activity. The concept of co-regulation has been extended by Panadero et al. (2019) as a mechanism that allows both student learning and the progression of teacher (or peer) interventions during formative assessment based on their interactions with students.

Research on student involvement in formative assessment in support of writing

Empirical research on student involvement in formative assessment has investigated two main topics: (1) students’ use of tools for self-assessment and peer assessment, in particular rubrics, scripts and checklists, provided by the teacher or by a researcher; (2) students’ participation in classroom discussions of assessment criteria and standards, embodied in assessment tools, as well as analysis of exemplars (model texts, texts constructed to exhibit shortcomings, sample student texts). Extensive reviews of research on these topics are presented by Andrade (2019), Andrade and Valtcheva (2009), Panadero et al. (2017), and Panadero et al. (2016). The research reviewed concerns formative assessment in a wide range of subject matter areas and educational contexts, with many more studies conducted in secondary school and higher education than in primary school. In a paper discussing students’ evaluative judgment in higher education, Tai et al. (2018) outlined a series of proposals for enhancing student involvement and sense of ownership of formative assessment practices. Their proposals include the following: feedback as dialogue rather than as correction; co-creation of rubrics to develop shared understanding; linking rubrics and exemplars; analysis of multiple exemplars to illustrate a range of quality indicators. All of these proposals are of interest but their applicability, particularly in primary school, still needs to be demonstrated.

The rest of this section is focused on findings of research on formative assessment of writing in primary school. In some cases, the publications also include data from the first years of middle school. Graham et al. (2015) carried out a meta-analysis of the effects of formative assessment on student achievement in writing in grades 1 to 8. The results showed weighted effect sizes of 0.87, 0.58, and 0.62 respectively for formative feedback about writing from adults (primarily teachers), from peer assessment, and from self-assessment. The authors warned against overinterpretation of the differences between the effect sizes since the different forms of feedback were not directly compared in most of the studies reviewed. The findings of this meta-analysis are coherent with more general findings on the positive effects of formative feedback (Hattie & Timperley, 2007). It should be noted, however, that most studies estimated the effects of feedback on outcome measures without providing process-tracing data on how the effects occurred: that is, without indications linking specific aspects of feedback to specific improvements in student writing. How students use feedback – what actions they take or fail to take after receiving feedback – needs more systematic investigation according to a review by Van der Kleij and Lipnevich (2020) of research on feedback from assessment.

A closer look will now be given to empirical studies of student involvement in formative assessment of writing in primary school classrooms. With respect to self-assessment, two studies are of particular interest because students not only used self-assessment tools but also made some contributions to the formulation of these tools. In a study by Ross et al. (1999) on narrative writing in grades 4, 5 and 6, the teachers in the treatment condition guided students in brainstorming about criteria, incorporated student proposals and language into assessment rubrics, supervised students’ use of the rubrics for self-assessment of writing and the formulation of writing goals. The average effect size was quite small (0.18), but the effect size for writers with weak skills was substantial (0.58). For these students, improvement occurred for the integration of story elements around a central theme and for the adoption of a narrative voice, but no improvement was found for the use of language conventions (grammar, spelling).

Andrade et al. (2008) studied the effects of self-assessment based on rubrics for writing stories or essays in grades 3 and 4. Students in the treatment group read a model story or essay and generated criteria for an effective story or essay. Elements of these criteria were then incorporated into a rubric formulated by the researcher and given to the students for self-assessment and revision of their texts. When using the rubric, students underlined with coloured pencils key phrases in the criteria presented in the rubric (e.g. for essay writing, ‘states clearly an opinion’) and corresponding elements in their drafts (e.g. words expressing his or her opinion). This self-assessment process was carried out for all criteria (Ideas and Content, Organisation, Voice, Word choice, Sentence fluency), except Conventions (spelling, punctuation, capitalisation). Self-assessment using the rubric led to positive improvements in subsequent writing with respect to higher-order aspects of writing (Ideas and Content, Organisation, Voice, Word choice), but not with respect to Sentence fluency and Conventions.

Studies of peer assessment and collaborative peer revision in primary school have shown that these activities can help students improve their texts. For example, Rouiller (1996) compared reciprocal peer revision and individual author revision on drafts of narrative writing in sixth grade. Peer revision had a positive effect in terms of number and quality of revisions, but the effect did not transfer to subsequent individual writing of a new text. Several studies have indicated that peer assessment and revision are likely to be more effective if they are coupled with strategy instruction or other preliminary forms of guidance. Olson (1990) compared four conditions in a study of sixth-grade students writing autobiographical stories. The condition combining reciprocal peer revision and direct instruction in revision techniques led to higher quality writing on a post-test text than the other conditions (direct instruction alone, peer revision alone, control condition without either direct instruction or peer revision). Direct instruction was carried out in five lessons that taught specific revision tactics (adding, deleting, substituting, paraphrasing, rearranging) and that emphasised different ways of carrying out these operations. In a study conducted by Boscolo and Ascorti (2004) in grades 4, 6, and 8, students individually practiced revision of a proposed narrative text containing various problems of cohesion (information gaps, inconsistencies). They then wrote personal narratives under two conditions: reciprocal peer revision, or corrective feedback from a teacher. On a subsequent narrative text written individually (post-test), students in the peer revision group produced texts with fewer information gaps. Analysis of the interactions of the students during peer revision showed that in primary school (grades 4 and 6) students principally expressed requests for clarification or for missing information, whereas in grade 8 they made a relatively larger number of suggestions about how to improve the text. Studies carried out with students with learning disabilities in the upper primary grades have also shown that strategy instruction combined with interactive peer revision was more effective than peer revision alone with respect to improvement of text quality (MacArthur et al., 1991).

A study by Meusen-Beekman et al. (2016) in grade 6 used a different type of outcome measure: namely, self-report questionnaires concerning self-regulation, external regulation, intrinsic and extrinsic motivation, and self-efficacy. Two types of student involvement in formative assessment (self-assessment and peer assessment) were compared to a control condition across three essay writing assignments The results showed that self-assessment and peer assessment had positive effects on measures of self-regulation and intrinsic motivation, but no significant effects on the other measures. Self-regulation and peer regulation had equivalent effects.

The research presented in the next section concerns a writing activity that combines several forms of student involvement in the co-construction of formative assessment processes.

A study of student involvement in formative co-regulation of writing

This section first summarises the characteristics of a writing activity designed to promote student involvement and cites key findings from a study of its implementation, as reported in more detail in a previous publication (Allal, 2018). It then presents a series of qualitative observations collected during this study, but which were not previously reported. These observations provide concrete illustrations of ways in which students can actively contribute to the process of co-regulation through participation in interactive formative assessment processes and in the co-construction of a formative assessment tool to use when drafting and revising their own texts (self-assessment) and when discussing with another student about possible revisions of their respective texts (reciprocal peer assessment) .³ Implementation of the activity was studied in three classrooms characterised by very similar teacher profiles and student populations.

The writing activity, entitled ‘The Life of a Star’, took place during fifth grade and then again in sixth grade with the same teachers who continued with the same classes. The activity required students to choose a star (in the area of music, sports, cinema, etc.) and write texts, as if they were the star, in response to questions from a journalist for an article to appear in a magazine. The journalist’s questions were:

1. When and how did you begin to get involved in your activity?

2. What was the most outstanding success of your career and why?

3. What problems do you encounter as a star?

4. What advantages do you have as a star?

The writing activity consisted of three sessions organised on three consecutive mornings:

• Session 1: Students read examples of magazine articles of interviews with stars and discussed features characteristic of this text genre; dyads were formed and each dyad chose a star (a person or a group) to write about.

• Session 2: The teacher led a whole-class discussion about possible answers to the journalist’s questions, followed by the co-construction with students of a Writing Guide and interactive practice using the guide to revise a sample sentence; each student then produced a draft text.

• Session 3: Using the Writing Guide and other reference materials, students marked proposed revisions on their partner’s draft and on their own draft; each dyad then discussed the proposed revisions and each author decided on the revisions to be retained in his or her final text.

The design of the activity was based on the previously mentioned model of co-regulation in which formative assessment is embodied in interactive processes (teacher-student exchanges, peer interactions) and in tools aimed at enhancing these processes as well as those of student self-regulation (Allal, 2018).

Observations in the classrooms during Session 2 showed that the discussion of potential content and the co-construction of the Writing Guide involved more active student participation and led to a more elaborate guide in one class (taught by Ms. Barrow) than in the other two classes. The frequency and types of revisions made by the students in Session 3 were analysed by means of chi square tests and repeated measures analyses of variance. The main findings from these analyses were that the students in Ms. Barrow’s class made a significantly larger number of revisions and, more importantly, carried out more extensive higher-order revisions (enrichment of text content and transformation of text organisation) than the students in the other two classes. It was also found that the investment of Ms. Barrow’s students in rewriting (adding content, organisational transformations) did not lead them to neglect the correction of spelling errors (i.e. their final texts did not contain a higher proportion of grammatical and lexical spelling⁴ errors than those of the students in the other two classes). In addition, the analyses showed that both self-regulation (revisions made individually by the author of his or her text) and peer regulation (revisions based on interactions between peers) contributed to improvement of the students’ final texts. Around 83% of the revisions from each source had positive outcomes: namely, accurate correction of errors, enrichment of text content, and improvement of text organisation in line with the items in the Writing Guide. Given these findings, it is of interest to look more closely at observations collected in Ms. Barrow’s class regarding the ways student involvement was fostered and some of the consequences.

The following two subsections present qualitative data collected in Ms. Barrow’s classroom. The first subsection is based on observations recorded during the writing activity and on notes of discussions with Ms. Barrow. Its aim is to provide insights regarding the features of the whole-class discussions that fostered active student involvement in the co-construction of formative support for writing. The second subsection is focused on observations of a student with weak basic skills in writing. Its aim is to allow critical reflection about the benefits and the limitations of interactive formative assessment.

Key features of whole-class interactive regulation in Ms. Barrow’s classroom

Two aspects of the way Mrs. Barrow conducted whole-class discussions remained constant between fifth and sixth grades and thus reflected basic principles of her approach to implementing interactive formative assessment in support of student writing.

Brainstorming about potential content

Ms. Barrow explained to her class that the suggestions all students make will contribute to a pool of ideas that each student is then free to use in his or her text. In response to a student’s question, she answered: ‘Yes, you are free to invent the star’s answers. If you know something, you can use it, but if you don’t know, you can invent.’

In contrast to the other classes where the teacher immediately launched a whole-class discussion, Mrs. Barrow had the students first discuss the journalist’s questions in small groups (3–4 students) as a way of preparing the whole-class discussion. The aim of this strategy was twofold: (1) increase the opportunity for all student to express their ideas in the small-group setting; (2) increase the probability that students who do not generally intervene in whole-class discussions will be stimulated (having tried out their ideas in a small group) to make proposals in the whole-class setting.

Here are two excerpts of student proposals in the whole-class discussion in sixth grade.

Question 2. What was the most outstanding success of your career and why?

‘My first album’

‘I received lots of fan mail’

‘When I was invited to sing at Johnny Halliday’s birthday '

‘When I saw my mother weep at my concert’

‘My fans threw lots of things at me’

‘I saw my face in all the magazines’

‘When I had my first platinum record’

Ms. Barrow: ‘And if the star is in sports?’

‘When I received my 7^th gold medal’

Question 3. What problems do you encounter as a star?

‘Receive unwanted letters and not have time to open them’

‘Always signing autographs’

‘Paparazzi follow us’

‘Fans touch us’

‘They tap on the window of the restaurant where we are eating’

‘A burglary: because I’m rich and they want to take something personal’

Ms. Barrow: ‘What personal thing?’

‘A photo with a boyfriend’

As these illustrations show, Ms. Barrow generally refrained from making evaluative comments indicating approval or disapproval of students’ ideas but asked questions that could enlarge the range of ideas or allow exemplification.

During the brainstorming about content, Ms. Barrow wrote on the blackboard vocabulary used by the students that is specific to the type of text to be produced and may not be known to all students. For example, in fifth grade, the list included: an autograph, a casting, the fans, the paparazzi, a faked photo, an interview, Olympia (name of a famous musical venue in Paris).

Co-constructing the Writing Guide

The purpose of the Writing Guide was to provide a list of indicators about the qualities of the text to produce so that students could refer to these indicators while drafting their texts and also during the revision activities (author revision of his or her text; reciprocal peer revision and discussion). The guide was constructed primarily on the basis of proposals from the students which the teacher wrote on the blackboard. In contrast with the other classes, Ms. Barrow formulated the items in terms of the student’s actions as a writer (e.g. ‘I write events in chronological order’.). Table 1 presents the Writing Guide constructed in sixth grade. Each of the first six items in the guide was proposed spontaneously by a student, but Ms. Barrow asked follow-up questions to solicit elaboration, for example:

A student proposes:

‘I adopt the viewpoint of the star’

Ms. Barrow:

‘What does that mean?’

Student:

‘I talk about myself’

Ms. Barrow writes on the blackboard:

I use ‘I’

She then asks:

‘And what if you are a group?’

Several students:

‘We’

Ms. Barrow writes:

or ‘we’

Table 1. Co-constructed tools displayed on the blackboard during the writing activities in sixth grade

Writing Guide for ‘The Life of a Star’ What do I need to think about? - I avoid repetitions - I adopt the viewpoint of the star - I use ‘I’, or ‘we’ - I check spelling & grammatical agreements (*) - I write events in chronological order - I put punctuation (periods, commas) - I use transition words . temporal (in the past, several years ago, today, when, suddenly*) . logical (because, since, that’s why, it’s thus/then, undoubtedly*) - Q 1–2: I write in past tense (perfect, imperfect tenses) - Q 3–4: I write in present tense (or past**) (*) *The students gave various examples orally. **The teacher indicated that when remembering an earlier event, past tense can be used.

Revision of the sample sentence fabulous     when I’m in the street, After this/ experience/, something that always frightens me is when dozens of excited fans run towards me asking me for an autograph. shouting and applauding … . and start photographing me In addition to additions and reformulations (in bold above), the students proposed corrections concerning lexical aspects of spelling, grammatical homophones, plural markers, noun-adjective agreement, subject-verb agreement, capitalisation of first word in the sentence.

The last two groups of items in the Writing Guide (transition words, verb tenses) were initiated by questions from Ms. Barrow, but the students actively participated by giving examples of transition words and of expressions with the verb tenses. Although the item ‘I check spelling & grammatical agreements’ is expressed in a general way, it is a reminder of activities that the students in this class previously carried out concerning the choice of an appropriate reference (e.g. dictionary, conjugation tables, etc.) for checking different aspects of spelling and the techniques to use for verifying grammatical agreements and for differentiating grammatical homophones.

In order for the students to practice using the Writing Guide while revising a text, Ms. Barrow wrote a lengthy sample sentence containing a number of errors on the blackboard. The errors concerned lexical aspects of French spelling, grammatical aspects (grammatical homophones, plural markers, noun-adjective agreement, subject-verb agreement), as well as punctuation. Ms. Barrow asked the students for corrections. They found all the errors and proposed accurate corrections that were marked in colour on the sample sentence. Ms. Barrow then emphasised that revision is not limited to correcting errors but includes enriching the content of the text to make it more interesting or making reformulations that improve the text. She asked the students to suggest examples of additions or reformulations for the sample sentence and added their suggestions on the blackboard (see the second part of Table 1).

A focus on one student: Vera

Vera and her family arrived in Geneva from a non-Francophone country two years before the research conducted in Ms. Barrow’s class. After two years in primary school classes of the Geneva public school system, Vera’s oral expression in French had greatly improved, but when writing texts, she still encountered significant difficulties with respect to both the lexical and grammatical aspects of French spelling and she sometimes showed misunderstanding of vocabulary usually mastered by Francophone students her age.

During the whole-class discussion in fifth grade, Vera was an active participant, repeatedly raising her hand to make suggestions. For two of the questions under discussion, her proposals led Ms. Barrow to provide interactive guidance:

• During the discussion of question 1 (When and how did you get involved in your activity?), Vera intervened saying: ‘She started singing when she was three’. Ms. Barrow asks her to reformulate as if she is the star, and Vera did so (‘I started singing when I was three’).

• During the discussion of question 4 (What advantages do you have as a star?), Vera suggested ‘One’s really tired’. Ms. Barrow asked her: ‘Is that a plus in life?’ (citing the definition of advantage given by another student). Vera shook her head ‘No’. Ms. Barrow asked her: ‘Can you give an advantage in other words?’ Vera answered: ‘It’s everything one can have: houses, cars … ’ Ms. Barrow suggested that Vera make a note of her ideas for her draft.

For Ms. Barrow, it was important to provide (in her words) ‘mini-scaffolding’ to help students formulate a more adequate idea, rather than corrective feedback that could be demotivating.

In the final version of Vera’s text in fifth grade, there were several traces of the influence of the whole-class discussion (see Table 2). Vera writes as if she’s the star (Question 1) and elaborates on advantages she mentioned briefly during the interaction with Ms. Barrow (Question 4). Vera also carries out substantial rewriting aimed at enriching the content of her text (addition of 43 words) and makes other revisions in line with the Writing Guide (e.g. addition of a transition word, correction of spelling errors). Her final text contains, however, a large number of errors: 43 incorrectly spelled words in a text of 162 words (26.5%).

Table 2. Excerpts from Vera’s text in relation to the whole-class discussion and the Writing Guide

Vera’s final text	Relations with the whole-class discussion and the Writing Guide
Question 1: Since the age* of 3, I started* to sing in front of my parents my friends, then in TV shows* … Bold: Words added during the revision session	She adopts viewpoint of the star. She enriches the content (my friends added). She adds a transition word (then).
Question 2: Addition of 43 words during the revision session (mentions her first offer from a record company and feeling very proud)	She enriches the content in a substantial way.
Question 4: I have the possibility* to buy myself 3 houses … luxury* cars, chic* clothes* Bold: Words added, and correction made during the revision session	She elaborates on the advantages she had mentioned orally. She corrects a reflexive verb construction (magiter changed to m’acheter)
*words spelled incorrectly in French

In sixth grade, Vera’s revisions show her continuing interest in enriching the content of her text by adding content or rewriting passages. She also shows progress in the appropriate use of transition words, both temporal (when, then, afterwards, now) and logical (because, so that). Her adoption of the voice of the star being interviewed is audible in several expressions: ‘And one fine day like a miracle … ’; ‘Well I couldn’t do anything, that’s life … ’; interjection of laughter: ‘ha, ha, ha’. Her final text contains relatively fewer errors than in fifth grade (34 incorrectly spelled words in a text of 262 words: 13%). This is a sign of progress, but also an indication of a persisting problem that may affect her future options as she enters secondary school.

Discussion of the observations from Ms. Barrow’s class

A study comparing three classrooms carrying out the same writing activity (Allal, 2018) showed that in the classroom of one teacher (Ms. Barrow) the students participated more actively in the co-construction of formative assessment processes and subsequently made more improvements of their texts (more revisions with a positive outcome and a larger number of higher-order revisions) than in the other two classrooms. The qualitative observations presented in this paper provide indications regarding the key features of Ms. Barrow’s strategy during whole-class discussions. These features may be summarised as follows:

1. Have students read examples of the text genre to be produced so that they have a basis for identifying possible criteria to guide drafting and revision.

2. For different topics to be dealt with in preparation for writing, alternate between whole-class discussion and small-group exchanges so that more students actively participate in the brainstorming about content and criteria.

3. Include, in the whole-class discussion, remarks to individual students that provide ‘mini-scaffolding’ for the subsequent drafting of their texts.

4. Make constructive links, during the whole-class discussion, between the different contributions expressed by the students.

5. Involve students in the co-construction of a formative assessment tool to be used for drafting and revising their texts and display the outcome (e.g. on the blackboard) to promote collective ownership.

6. Provide students with an opportunity to practice using a formative assessment tool on exemplars (or excerpts) and display the outcomes (e.g. examples of various revisions).

7. Promote students’ use of a formative assessment tool for self-assessment and peer assessment when reviewing and revising their texts.

The above principles coincide with several points identified by Ruiz-Primo (2011) in her analysis of instructional dialogues as a means of informal formative assessment in whole-class discussions. In addition, they overlap with aspects of the writing instruction strategies investigated by Andrade et al. (2008), Boscolo and Ascorti (2004), and Ross et al. (1999), while providing more detailed indications about how a teacher may encourage student involvement in the context of whole-class discussions. It is important to understand how processes of formative co-regulation can be integrated within whole-class discussions because these activities often set the stage for the subsequent writing activities (individual drafting and revision, peer review and revision, etc.).

Ms. Barrow’s interventions while the students were working individually or in dyads were limited to reminding students to refer to the Writing Guide. More intensive co-regulation could, however, occur if the teacher interacted with individual students while they are drafting and revising their texts. Several researchers have provided detailed transcriptions of teacher-student dialogues in one-on-one and small group settings illustrating processes of co-regulation. In a qualitative study of one-on-one interactions between teachers and students, Heritage (2016) presents several transcripts, including one from a grade 5 writing classroom working on argumentative text. Her analysis of the transcript shows that co-regulation of student writing results from both the teacher’s actions (questions the teacher asks, types of feedback she gives) and the initiatives taken by the student (questions soliciting feedback, expression of his intentions and concerns). A monography by Torrance and Pryor (1998) presents an in-depth qualitative study of formative assessment in classrooms with students aged 4–7. Several transcripts illustrate the Vygotskian principle that scaffolding is effective when the interventions of each actor (teacher, student) appropriate elements of the interventions expressed by the other actor.

The observations presented in this paper concerning Vera show similarities to findings reported in two studies of formative assessment of writing in the primary grades (Andrade et al., 2008; Ross et al., 1999): namely, that student involvement in formative assessment processes helped writers with weak basic skills to improve important aspects of their writing (content, organisation, voice) but did not necessarily improve their mastery of conventions (grammar, spelling, punctuation). Students like Vera are often overwhelmed by the conventional aspects of writing and see revision as purely a process of correcting errors. If participation in the co-construction of formative assessment can help them engage with higher-order aspects of writing, this is to be applauded. At the same time, it is likely that formative assessment tools designed by experts in the field of language conventions will need to be integrated in classroom practice to allow significant progress in these areas. One question to explore in future research is how to combine co-constructed formative assessment tools and expert-designed formative assessment tools that, if well coordinated, could significantly improve student writing in the primary grades. Higher-order aspects of writing have some degree of generality for a given text genre (e.g. narrative schema, autobiographical voice) across languages. In contrast, linguistic conventions are very language specific. This means that expert-designed tools focussed on language conventions need to be developed in each linguistic community.

Two limitations of the approach in Ms. Barrow’s class need to be mentioned. Her technique of asking for students’ proposals for the Writing Guide and writing them on the blackboard in the order they were proposed led to a non-structured checklist of items. The school curriculum for writing generally provides conceptual categories that could be used in the co-construction of a Writing Guide or other types of formative assessment tools. For example, the teacher could display key categories (e.g. on the blackboard) and then ask students to suggest items and to indicate the category under which to insert each item. With respect to categories in the Geneva curriculum for upper primary grades, this could have led to a Writing Guide in sixth grade in Ms. Barrow’s class presented as follows (using the items in Table 1):

Text genre, voice, and content

- I adopt the viewpoint of the star

- I use ‘I’, or ‘we’

(+ brainstorming about content and specific vocabulary written on the blackboard)

Text organisation

- I avoid repetitions

- I write events in chronological order

- I use transition words

. temporal (in the past, several years ago, today, when, suddenly)

. logical (because, since, that’s why, it’s thus/then, undoubtedly)

- Q1-2: I write in past tense (perfect, imperfect tenses)

- Q 3–4: I write in present tense (or past)

Conventions

- I check spelling & grammatical agreements

- I put punctuation (periods, commas)

If formative assessment tools for different writing activities are co-constructed with the same categories (in bold in the example above), this could help students develop a general understanding of the main dimensions that need to be considered when drafting and revising any type of text.

A second limitation concerns the fact that the students’ practice using the formative assessment tool (Writing Guide), prior to drafting their own texts, was limited to the interactive revision of a long sentence written on the blackboard by Ms. Barrow. Interactive revision, guided by the teacher, of a complete text, or of several different exemplars of texts (as suggested by Tai et al., 2018), could have a larger impact on students’ understanding of criteria and their ability to make various types of revisions. However, since the time available for writing activities is never unlimited, teachers will need to find the right balance between the amount of time students spend practicing revision on exemplars furnished by teachers and the amount of time they are engaged in drafting and revising their own texts or reviewing the texts produced by peers.

One further comment: The Writing Guide presented above differs from rubrics used in other research on writing in primary school in two respects. The rubrics in the Andrade et al. (2008) study, for example, included 6 categories (Ideas and content, Organization, Voice, Word choice, Sentence fluency, Conventions) crossed with 4 levels of evaluation. Although the rubrics included some elements from student brainstorming about criteria, they were essentially researcher-designed instruments. The Writing Guide described above is less detailed (only three categories with items listed for each), but it retains the words students used in formulating the items. It can be argued that these characteristics are likely to facilitate active use by primary school students during the processes of drafting and revising their texts. There is, however, a lack of research comparing different formats of formative assessment tools for self-assessment and peer assessment of writing in primary school. Comparisons of this type could be a goal for future research.

Concluding remarks

Implications for teacher professional development and for future research are considered in these concluding remarks.

How should professional development activities for primary school teachers concerning formative assessment of writing be designed? These activities obviously need to take into account general findings from research on writing instruction and from investigations of formative assessment practices. The guidelines derived from research on self-assessment and peer assessment are particularly relevant but are often framed in very general terms (e.g. ‘Provide quality peer-assessment training, examples and practice…’, Panadero et al., 2016, p. 322). Qualitative observations of classroom practices in the area of writing, like the ones presented in this paper, need to be collected in order to assemble progressively a corpus of detailed examples that could be used to help teachers think concretely about how to encourage student involvement in formative assessment of writing.

A professional development workshop on formative assessment of writing in the primary school classroom could unfold as follows. To start with, participating teachers would be asked to talk about formative assessment practices in the area of writing they have tried and the benefits and/or drawbacks they have observed. Case studies of several different strategies developed by teachers could be presented and discussed. Participating teachers could then try out one strategy or another and share their observations. In professional development that is school-based, or integrated in a network of practitioners (Garet et al., 2001), teachers could decide on a common strategy or decide to try out different strategies and compare their respective observations. Professional development must incorporate research findings as they become available but has to move forward in response to teachers’ needs and initiatives, even when there are no definitive answers from research to many questions.

More research is obviously needed to understand both the effects of active student involvement in formative assessment of writing in primary school and the processes underlying the effects. Three directions of research need to be developed. The first is the design of studies that coordinate the measurement of the effects of different aspects of student involvement with the collection of process-tracing data on how student involvement occurs, its dynamics, benefits, and limitations. The second is the comparison of different types of formative assessment tools (provided to students vs. co-constructed with students) as used by primary school students while they are drafting and revising their texts. The third is the investigation of ways of articulating formative assessment tools co-constructed with students with expert-designed tools aimed at improving specific skills that are not sufficiently enhanced by co-construction. The design of these types of investigation will not be a simple matter. It will require a programmatic and interdisciplinary approach involving specialists in formative assessment, writing instruction, and linguistics. In addition, because most writing instruction activities are complex and have several interdependent components, it is often difficult to isolate specific effects or specific processes. Researchers will need inventiveness and persistence to move the field forward, and they will need to do so in collaboration with teachers. A review of research on ‘researcher-practice partnerships’ (Welsh, 2021) presents the benefits that can result from these partnerships and also the challenges faced on both sides. The description by Jessen and Parr (2019) of several researcher-practitioner partnerships in the area of writing highlights the contextual factors that influence research design and implementation. Contextual factors, such as the assessment policy of the education system, the writing curriculum, the content and organisation of teacher education, the linguistic and cultural background of the students, all need to be taken into account in designing a partnership. Nevertheless, it can be argued that the sharing of innovative ideas and empirical evidence from research across contexts is essential for the development of more effective approaches to writing in each context.

Acknowledgments

I thank the members of the team who worked with me over a number of years on research concerning writing in primary school classrooms and who made valuable contributions to the work referenced here: Céline Buchs, Jamila Dorner, Katia Lehraus, Lucie Mottier Lopez, Walther Tessaro, and Edith Wegmuller. I am also grateful to the teachers and the students who participated in the two-year study.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Notes on contributors

Linda Allal

Linda Allal received a Ph.D. in Educational Psychology from Michigan State University, USA, in 1973, and was awarded an honorary doctorate from the University of Liège, Belgium, in 2013.After a career spanning 33 years at the Faculty of Psychology and Education Sciences of the University of Geneva, Switzerland, she is professor emeritus since 2006. Her research concerns the relations between learning, teaching, and assessment in school settings. Her recent publications address issues of assessment for learning, processes of formative assessment in the co-regulation of classroom writing activities, and teachers’ professional judgment in their practice of summative assessment.

Notes

1. This paper examines assessment of writing in regular primary school classrooms (through grade 6, approximately age 11–12) in activities using the primary language of instruction of the school system (L1 writing). The classrooms considered are not geared specifically to second language learners or to students with special needs, although some students in such classrooms may have another home language than the language of instruction and/or may encounter various learning difficulties.

2. The expression ‘assessment for learning’ is increasingly used to encompass formative assessment and some aspects of summative assessment that provide positive support for student learning (Laveault & Allal, 2016). The present paper on formative assessment of writing includes contributions from the assessment for learning literature, without reviewing the definitional issues discussed in the above book chapter.

3. The description of the writing activity and the qualitative data presented here have been translated from French to English by the author. Fictive names have been given to the teacher and the student who are cited.

4. French spelling includes lexical aspects (how individual words are transcribed) as well as many complex grammatical aspects that concern agreements in noun clauses (between articles, qualifiers and nouns), subject-verb agreements, past participle agreements with antecedents, differentiation of grammatical homophones, verb tense coordination, among other aspects.