“What are they talking about?” A sociocultural linguistic approach to practical task effectiveness

Abstract

This case study demonstrates teaching and learning activities in the school laboratory, and employs talk moves for the direct assessment of practical task effectiveness. By adopting a sociocultural linguistic approach (SCLA), learning chemistry is understood to be a discursive process in which knowledge is constructed through social interaction and language. Thus, learning may be identified by attending to the language used in classroom discourse. The multimodal communication that took place during two acid and alkali practical lessons for learners aged 11 and 12 years was filmed and transcribed. Analysis of the transcripts revealed the language learning opportunities afforded by the tasks and demonstrated that school chemistry practical lessons can be understood in terms of three linguistic opportunities: introducing, using, and reflecting upon language. This lesson structure could be employed to plan more inclusive and equitable practical lessons which foreground language and value discussion equally to manipulating equipment. Recasting practical lessons as sites for learning and using the language of chemistry, key words introduced by the teacher are tracked and counted throughout the lesson to identify when they are used and by whom. The novel 3-part practical (3P) framework and multimodal discourse analysis are employed to assign the use of key words to the macroscopic, submicroscopic or symbolic level of thought. This analysis reveals the centrality of a results table to structuring talk and the detrimental effect of introducing novice learners to multiple levels of thought simultaneously. The Talk Identification (ID) Grid has been developed and used here to analyse student group discourses using talk moves to signpost learning in the domain of ideas and the domain of observables. Descriptors are provided to support instructors in identifying talk moves and how these moves relate to practical task effectiveness to target interventions that improve learning procedural and conceptual knowledge in the laboratory.

---

Introduction

Language is the principal medium through which teaching and learning occurs enabling learners to demonstrate their knowledge and understanding. Studies in chemistry education report that language is a barrier that impacts learning, participation, and attainment (Cassels and Johnstone, 1984; Byrne et al., 1994; Markic, 2015; Markic and Childs 2016; Rees et al., 2018, 2021).

There is an acknowledged understanding in education that all students are language learners, and all teachers are language teachers (Bullock, 1975; de Oliveira, 2016) many teachers do not feel able to provide language support in their lessons beyond teaching scientific vocabulary (Markic, 2015; Quílez, 2021). However, vocabulary alone is not enough; instead students need to learn both words and how to use them (semantic structures) if they are to make the same meaning as their teachers (Lemke, 1990). Recommendations for teaching science to English language learners include: providing opportunities for productive discourse and interactions with others; using multiple modalities; and engaging in disciplinary practices (National Academies of Sciences, Engineering, and Medicine, 2018). Such opportunities are provided by hands-on practical work when students are given the time to discuss their work, and to reflect upon their understanding of the pertinent scientific terms and concepts (Lemke, 1990; Tobin, 1990; Lunetta et al., 2007; Abrahams and Reiss, 2012; Gatsby, 2017). This can be considered through the lens of sociocultural linguistics.

Sociocultural linguistics

Sociocultural linguistics is the broad interdisciplinary field concerned with the intersection of language, culture, and society, and encompasses sociolinguistics, linguistic anthropology, and linguistically oriented social psychology, among others (Gee, 2008). A sociocultural linguistic approach emphasises the importance of social interaction and language for thinking and learning (John-Steiner and Mahn, 1996).

Language and learning are inextricably linked (Halliday, 2004). Chemical concepts, as an example, do not exist in the abstract but are constructed by language blended with multimodal communication (O’Halloran, 2005, 2015; Gilbert, 2010), where multimodal is used here to refer to semiotic modes such as language, image, and gesture, rather than perceptual modes such as visual or haptic (Silliman et al., 2018).

Understanding the centrality of language in chemistry education substantiates the adoption of sociocultural linguistic approach and the application of language teaching and learning approaches in school chemistry. Understanding is challenged through exploratory talk (educationally effective talk) creating new knowledge, facilitating students to work in their zone of proximal development (Vygotsky, 1978, p. 86). Student talk can be scaffolded using specific talk moves to develop discussion (Chin, 2006; Michaels and O’Connor, 2012). Scaffolds may be pre-planned macro-scaffolds or spontaneous micro-scaffolds (Nielsen and Hougaard, 2018). Talk moves are conceptualised as tools for facilitating academically productive talk (Michaels and O’Connor, 2015) and are used here to refer to talk occurring between teacher-student and student–student. Following from the work of Andersson and Enghag (2017), but in a school chemistry context, talk moves are conceptualised as tools that signpost how students talk during practical tasks. By identifying how students talk, targeted interventions may be implemented to better facilitate exploratory talk.

Language, cognition, and communicating chemistry

Language is conceived by Vygotsky as both a social instrument and a psychological tool (Mercer, 2002). Words and thoughts are inseparable, “thought is not merely expressed in words: it comes into existence through them” (Vygotsky, 1987, p. 219). Chemistry education requires “multilevel thought” (Johnstone, 1991 p.78), specifically three levels of thought which are often conceived as the apices of a triangle. These are the macroscopic (macro), observable and tangible phenomena that can be experienced with our senses; the submicroscopic (sub-micro), models of matter, atoms, molecules, ions, and structures that have descriptive or explanatory roles; and the symbolics which include all the chemical and mathematical signs and images used to represent chemical concepts (Johnstone, 1982; Johnstone, 1991; Talanquer, 2011).

To be successful, learners of chemistry must be able to make connections among these varied representations (Yore and Treagust, 2006), but teachers should focus on one level of thought at a time to secure students’ understanding (Georgiadou and Tsaparlis, 2000; Tsaparlis et al., 2010).

Cognitive load theory explains that all tasks place a demand on the participant's working memory (the intrinsic load), and when instruction draws on multiple levels of thought, a high (extraneous) load is also placed on the learner's working memory. The greater the extraneous cognitive load imposed, the fewer the cognitive resources available for dealing with intrinsic cognitive load and so less learning occurs (Sweller et al., 2019).

The way information is presented during laboratory work is of particular importance, because the tasks themselves are demanding and impose a high cognitive load on the participants’ working memory (Johnstone and Wham, 1979).

The purpose of hands-on practical work may be understood to link the domain of ideas and the domain of observables (Tiberghien, 2000). The domain of observables is used here to refer to procedural knowledge: what is done with objects; and making observations. In contrast, the domain of ideas considers conceptual knowledge: the theories; and ideas that underlie the activity. Student talk during laboratory work has been reported to focus on the procedures needed to carry out the experiment (Russell and Weaver, 2011; Sandi-Urena et al., 2011), which suggests that learners are “manipulating equipment and not ideas” (Hofstein 2017, p. 366). The models presented by Tiberghien and Johnstone have been combined here (Fig. 1) to identify thinking by attending to talk that occurs during practical lessons.

Fig. 1 Three levels of thought (after Johnstone, 1991, p. 78) aligned with the domain of ideas and the domain of observables (Tiberghien, 2000).

Content and Language Integrated Learning is a language teaching approach that employs the target language for teaching and learning the subject matter (Dalton-Puffer et al., 2010; Dalton-Puffer, 2011). In Content and Language Integrated Learning science education, a practical lesson is understood to be composed of three distinct parts, each of which provides distinct linguistic opportunities (Nikula, 2015). Firstly, the pre-experimental phase exposes students to the subject's concepts, specialised vocabulary, and grammatical structures. Next, the experimental phase affords learners the opportunity to use the language modelled by the teacher during the introduction. Finally, the post-experimental phase can include metalinguistic work, thinking about the language used and how it is used. Nikula's description of a three-part practical lesson and Fig. 1 have been combined to develop the Three-Part Practical (3P) framework presented in Fig. 2.

Fig. 2 The 3-Part Practical (3P) framework for learning the language of chemistry.

The 3P framework operationalises key words (used here to refer to subject-specific language and symbolics) as markers that identify the domain and levels of thought being used during each part of the lesson. Multimodal data is required to contextualise the term, as for example, the word water could be used in talk concerning the procedure as a liquid at the macro level or in conceptual talk as a particle or molecule at the sub-micro level, and the meaning of H₂O at the symbolic level. However, when the word is spoken by a person holding a measuring cylinder, the macro level is implied. The 3P framework will be used to evidence whether both domains are referred to, and whether different levels of thought are drawn upon.

The 3P framework provides a temporal view of language use during the lesson: key words introduced by the teacher can be tracked through the lesson to see who is saying what, and when. Understanding science requires more than knowing and using key words; learners must also use the same pattern of meaning relations (semantic structures) as their teacher to make the same meaning (Lemke, 1982). When a learner can use key words in the pattern that is valued by the scientific community, they demonstrate understanding.

Identifying and assessing learning during practical work

An effective practical task facilitates doing and learning in both the conceptual domain of ideas and procedural domain of observables (Abrahams and Millar, 2008; Millar and Abrahams, 2009). The problem with trying to assess learning and understanding is that they cannot be directly observed but must be inferred from a learner's response to a task or question (Millar, 2013). However, analysis of talk has been proposed to understand how the interaction and content of students’ communication is related to outcomes of their actions during physics practical tasks (Andersson and Enghag, 2017). As observation affords the direct assessment of students’ practical skills (Reiss and Abrahams, 2015), attending to student talk during practical tasks provides the opportunity to directly assess learning. The direct assessment of learning is operationalised here using a version of Andersson and Enghag's (2017) model, adapted here for chemistry and combined with Millar and Abrahams’ (2009) table clarifying the meaning of ‘effectiveness’ (p. 61). The resultant Talk Identification (ID) Grid and its component parts are available in the Appendix 1.

The exploratory talk moves shown in Fig. 3, the Talk ID Grid, are operationalised here as tools, or signposts, for identifying conceptual and procedural learning during the practical task. It is intended that the Talk ID Grid could be used as an assessment for learning (Black and Wiliam, 1998) tool to support teachers in the direct assessment of the effectiveness of a practical task as a site for communicating chemistry. The Talk ID Grid characterises talk moves so that chemistry educators can target interventions that develop the quality of student talk and facilitate conceptual and procedural learning.

Fig. 3 Talk ID Grid identifying exploratory talk moves in the domain of observables and in the domain of ideas at both level 1 linguistic, and level 2 cognitive to determine the effectiveness of a practical task for communicating (derived from Millar and Abrahams, 2009; Andersson and Enghag, 2017).

Purpose and research questions

This case study aims to contribute to the existing body of qualitative research concerning practical work in chemistry education by demonstrating how spoken language, within a multimodal context, may be employed to target interventions that develop teaching and learning. The following research questions (RQ) are addressed by attending to both the teacher and the students’ multimodal communication:

1. What kinds of opportunities do practical investigations afford students for learning the language of chemistry?

2. To what extent can student talk be used to determine a practical task's effectiveness as a site for communicating chemistry?

Methodology of research

A case study design is adopted here that incorporates multimodal ethnographic principles to afford the in-depth examination of two practical activities in the acid and alkali chemistry unit of work for learners aged 11–12, in an English secondary school. This descriptive case study is a single case with embedded units (Yin, 2003) as the same class is observed during two different practical lessons and their language use compared. Furthermore, the case may be understood to be instrumental (Stake, 1995, as cited in Baxter and Jack, 2008) as student talk is used to assess practical task effectiveness. Adopting a SCLA understands that knowledge is co-constructed by individuals in a social and cultural context through the medium of language (John-Steiner and Mahn, 1996). Learning is mediated by language thus, analysing classroom talk provides a mechanism for directly identifying instances of learning. The unit of analysis is turns, where a turn is the time when a speaker is talking (Coulthard and Condlin, 2014). Modes in addition to talk will be foregrounded or backgrounded to facilitate understanding these turns (Sacks et al., 1978).

Participants and recording

This work took place in an English Boys’ Academy. One class of 30 students aged 11–12 years and their chemistry teacher were observed studying the acid and alkali chemistry topic. The researcher, who teaches in the same department, observed two additional lessons prior to data collection, to trial equipment, and familiarise the participants with the research process. Then, two full 90 minute practical lessons named Litmus and Neutralisation respectively were observed and recorded (lesson objectives and method are available in Appendix 2). The students had been assigned seats at the beginning of the school year and habitually worked in groups with students seated next to them. The room layout determined the camera locations which in turn determined which groups were filmed. The same two groups of three students were filmed during both lessons.

The video recordings were taken by four cameras, one focused on the front of the classroom where the teacher habitually stands and demonstrates, two on benches positioned where the student groups carry out their practical tasks, and one accompanying the researcher as field notes were taken and students were interviewed. In total more than 10 hours of recordings were made. The video recordings from the two lessons were categorised by camera location and were transcribed verbatim. The data was cleaned and triangulated by the researcher by repeatedly watching the recordings and comparing the transcripts from different cameras to each other and to the field notes. The written work produced by each group was also collected and used to support and validate data from the recordings in the absence of a second researcher.

To preserve anonymity, students habitually seated on the bench with camera 1 are referred to as Group 1 collectively and S followed by a number (S1, S2, and S3) individually. Students habitually seated on the bench with camera 2 are referred to as Group 2 collectively and S followed by a letter (SG, SK, and SL) individually. The teacher is coded as T, and class members not in Groups 1 and 2 are assigned the generic code S.

The work was conducted in compliance with the British Education Research Association ethical guidelines (BERA, 2018) and British Association for Applied Linguistics (2021), aligning with the principles of informed consent, right to withdraw, and guarantee of anonymity. The school Headmaster acts as an overseer of all actions conducted in the school, and permission to complete this research was confirmed by him. All data collected and used during this research was securely stored and although permission was granted by participants and their caregivers, and all the images used have been treated to prevent the identification of participants.

Multimodal discourse analysis

A SCLA to discourse analysis understands learning is produced in linguistic interactions that employ a range of modes to make meaning, and so learning may be identified through the analyses of these linguistic interactions. Within this approach teaching and learning a chemical concept for example, is understood in terms of teaching language through which the concept is formed and learning to use this language in the same way. The first RQ is addressed by multimodal discourse analysis to examine how meanings are made during practical work using the 3P framework.

An inductive thematic analysis as described by Ary et al. (2018) of the transcripts was conducted to expose language learning activities in each lesson part. Identification and coding of classroom strategies including initiation response feedback (IRF) questioning, were informed by Lemke's guide for recognising teacher and student strategies of control (1990, Appendix 2).

After identifying and classifying all the spoken episodes, the original recordings of sequences of interest were revisited. The transcription of these sequences was expanded to include a multimodal analysis of corresponding actions and images. The multimodal analysis was informed by Bezemer and Mavers (2011) and Flewitt (2011).

Word frequency analysis was used to understand students’ language use in each of the three parts of the practical lesson using the 3P framework. Microsoft Excel was used for the transcription of talk from both practical lessons as demonstrated by Bree and Gallagher (2016). The discourse was separated into a turn per row of the spreadsheet and the search function was used to locate words of interest in the data. The key words were identified inductively through the transcription process. Once the term was located, the speaker and context were noted, and the frequency of use was calculated and tabulated in the 3P framework.

To address the second RQ, the transcripts of student-student dialogue as they perform the Litmus and Neutralisation lessons were deductively coded into one of three different talk types, disputational, cumulative, and exploratory talk (Mercer, 1995). Individualised decision making, and disagreement characterise disputational talk, the exchanges are short and composed of: assertions; counter assertions; competing; and defending. Cumulative talk is characterised by: instruction; repetition; confirmation; elaboration; and although positive, it is uncritical, so ideas are not challenged nor justified. In contrast, exploratory talk is both positive and challenging: criticism is both constructive and justified; opinions are sought, and joint decisions are made; everyone actively participates; the exchanges are longer and demonstrate reasoning.

The Talk ID Grid was then applied to assess the practical tasks’ effectiveness as a site for communicating the language of chemistry. A task that facilitates exploratory talk is more effective than one that does not, however, the extent of a practical task's effectiveness (Abrahams and Millar, 2008; Millar and Abrahams, 2009) may also be determined using the talk moves associated with conceptual and procedural talk.

Results and analysis

RQ 1: What kinds of opportunities do practical investigations afford students for learning the language of chemistry?

Identification of language learning opportunities in each phase of the practical lesson

The video-recording transcripts from the two lessons were categorised as pre-experimental, experimental, and post-experimental, depending on the activity taking place. Inductive thematic analysis was then used to identify language learning activities consistent with those described in the Content and Language Integrated Learning science three-part practical lesson model (Nikula 2015). Extracts of the multimodal transcriptions are available in the Appendix 3 with brief salient excerpts from the datasets detailed below.

Pre-experimental phase

Extract 1 from the Litmus lesson begins with the teacher showing the students a piece of apparatus and initiating dialogue by asking the class a question. A student volunteers an answer to which the teacher gives feedback. This Initiation-response-feedback (IRF) strategy encourages multiple responses, facilitates productive thinking, (Sinclair and Coulthard, 1975), and provides ongoing assessment to assist students to construct knowledge (Chin, 2006).

Litmus lesson pre-experimental phase (from Extract 1)

Teacher: what is it? [row 7]

Student: is it, eh, is it an acid waffle or something like that? [row 8]

Teacher: Shush, good idea but no. [row 10]

Student: Pallet [row 13]

The students suggest names consistent with similar looking more familiar items like “a pallet”. Through repeating the IRF sequence the teacher provides the students with time and opportunity to consider the apparatus directing his body and gaze to the student answering, and by doing so demonstrates that he values their suggestions. Barnes (2010), describes this process as “active learning” whereby ideas are shared and shaped between interlocutors, forging links between new and existing knowledge.

Extract 2 from the Neutralisation task also demonstrates language instruction, the students are exposed to key words (acid, alkali, neutral), both through the teacher's talk and a written procedure that is projected onto the board. Furthermore, the teacher's explanation of neutralisation has drawn upon the semiotics of chemistry (H⁺, OH⁻) in both the visible and auditory modes.

In the excerpt below, the teacher is standing next to the particle representations drawn on the board and is holding the conical flask containing the green neutral solution as he speaks.

Neutralisation lesson pre-experimental phase (from Extract 2)

Teacher: 23 drops boys, at this point it's neutral. [row 125]

Teacher: All the OH's have combined with the H's and made water, so in there now is just water. [row 126]

Teacher: Not acid not alkali because all the OHs and Hs have joined to make water. [row 127]

The teacher physically and verbally links the domains of observables and ideas, a “contextualization of concepts” (Jiménez-Aleixandre and Reigosa, 2006, p. 708). In doing the teacher is simultaneously drawing on all three levels of thought (Johnstone, 1991).

Experimental phase

The experimental phase of each lesson lasted approximately 40 minutes including the time required to pack away the equipment. Extract 3 is a transcription from the Litmus task which involves testing different solutions with red and blue litmus paper and recording and analysing the results in a table of results copied from the board.

The extract begins when Student 3 has just arrived and joins Group 1 as they begin testing their fourth solution, deionised water. In this excerpt, Student 2 directs Student 3 to test the solution first with red then blue litmus paper following the column sequence displayed in the results table.

Litmus lesson experimental phase (from Extract 3)

Student 2: litmus, red then blue. [row 136]

Student 3: Nothing happened, it's the same. [row 144]

Student 1: No change, it is not an alkali. [row 145]

When Student 3 reports the red litmus paper result to the group, Student 1 nods and then rephrases the observation in the manner previously modelled by the teacher. The student's talk and actions are directed by the results table, and they can express their observation using the teacher's language pattern.

The experimental phase of the Neutralisation lesson is noisy and unsettled. Extract 4 begins when Group 2 are adding sodium hydroxide solution dropwise to a conical flask containing hydrochloric acid and universal indicator. The student's need to count and record the number of drops added and the concomitant colour change.

Neutralisation lesson experimental phase (from Extract 4)

Student L: 20 [row 93]

Student K: 30 [row 94]

Student L: Oh yeah, write 30 turned orange. [row 95]

The students’ language is indexical and dependent on the details of the practical task with very little use of subject-specific language and no consideration of what the results may mean. There appears to be a lack of collaboration as Student L dominates both the talk and equipment. These students do not appear to have adopted the language modelled by their teacher when introducing the neutralisation task.

Post-experimental phase

The teacher-led post-experimental phase lasted approximately 15 minutes for each task, additional time was provided at the end of each lesson for the students to check their written work and ensure the room was left clean and tidy. Extract 5 from the litmus task begins when Student K from Group 2 volunteers to provide the result for the mystery solution that smells like vinegar.

Litmus lesson post-experimental phase (from Extract 5)

Student K: In red litmus no change. [row 92]

Teacher: So, what does that tell us? [row 93]

Student K: That there wasn’t, that it's not an [ac] alkali. [row 94]

Teacher: Good, it wasn’t an alkali. [row 95]

Student K: Observation with blue litmus, it turned red so,

it was an acid. [row 96]

The teacher has reinforced the requirement to test the solution with both red and blue litmus papers then record the results and analysis, by using an IRF sequence to model the language and thought pattern required to do so. The teacher gives verbal affirmation and repeats the student's analysis whilst recording the result in the table on the board. The feedback in this IRF sequence can be understood as a scaffold, specifically a spontaneous micro-scaffold, as the student is able to respond fluently with the next result and analysis.

In Extract 6 from the Neutralisation lesson, the teacher draws upon a range of modes to regulate difficulty and negotiate meaning during an IRF sequence.

Neutralisation lesson post-experimental phase (from Extract 6)

Teacher: If we’re weakly acidic what are we going to have more of? [row 14]

Student: H plus. [row 15]

Teacher: Yeah, we’re still going to have more H plus aren’t we. [row 16]

Teacher: I’m going to put extra H plus, just to mean that there is not as many as the excess H plus in the strong acid. [row 17]

The teacher uses gesture to draw the student's attention toward the particle diagram on the board to scaffold his question and the student successfully negotiates the teacher's meaning. Having elicited the student's response, the teacher then marks the importance of the point by confirming and then rephrasing the answer (Lemke, 1990).

Furthermore, the teacher critically reflects on language and engages in metalinguistic work by writing and talking about his choice of terms “extra” and “excess” to denote a change in magnitude.

In summary, although the Litmus and Neutralisation lessons were not planned as language lessons, there is evidence that the teacher and the students are involved with language teaching and learning activities consistent with the Content and Language Integrated Learning science three-part practical lesson model (Nikula, 2015).

Tracking the introduction and uptake of key words

Inductive thematic analysis was used to identify the key words introduced by the teacher in the pre-experimental phase of the lesson. These terms were then used to deductively analyse the full transcripts from the lessons. The frequency with which participants used these terms in each part of the lesson was calculated and tabulated for the Litmus lesson (Fig. 4) and Neutralisation lesson (Fig. 5). Incomplete terms such as, red used to refer to red litmus paper, have not been counted in the word frequency table.

Fig. 4 Litmus Task 3P Framework of key word frequency use during each phase of the lesson demonstrates that the students are working in the domains of observables and ideas.

Fig. 5 Neutralisation Task 3P Framework of key word frequency use during each phase of the lesson demonstrates that the students are not working in the domain of ideas.

The frequency of key word use shown in Fig. 4 and 5 indicate that teacher talk dominates the pre-and post-experimental phases of the lessons. As previously demonstrated in the pre- and post-experimental phase Litmus and Neutralisation lesson exerts above, the teacher controlled the talk by selecting respondents one at a time, whereas during the experimental phase up to six students were talking, thus the word frequency values increased. The use of the word acid for example, is used by Student 1 in the Litmus lesson experimental phase (Extract 3) and by two other students in the Litmus lesson post-experimental phase (Extract 5). However, it is not always possible to assign the use of a word to a particular student during group interaction.

The Litmus lesson data presented in Fig. 4 shows an increase in frequency of students using key words from the pre-experimental phase to the post-experimental phase which may indicate that the practical task has increased students’ familiarity and confidence in using the terms as learning to apply words correctly facilitates understanding their meaning (Toulmin, 1972).

The Litmus lesson (Fig. 4) demonstrates 78 incidences of key terms being used by the students during the experimental phase but only 25 are recorded during the Neutralisation lesson (Fig. 5). These results indicate that the Neutralisation task is less effective than the Litmus task at facilitating students to adopt the language modelled by their teacher during the pre-experimental phase. For example, Fig. 5 records one incidence of a student using the word neutral, which in occurred during the Neutralisation lesson pre-experimental phase (Extract 2, row 124).

Identifying the key word use in both domains and at different levels of thought

Multimodal data recorded in the transcripts was employed to provide the context in which a key word was being used. In Fig. 4 from the Litmus lesson, the word acid was coded to both the domain of observables and the domain of ideas, corresponding to the macro and sub-micro levels of thought respectively. For example, in the post-experimental phase litmus excerpt above (from Extract 5), Student K states “Observation with blue litmus it turned red, so it is an acid.” The word acid is coded here in the domain of ideas because the student is drawing a conclusion based on the observation that the litmus has changed colour. In this context the word acid is understood to correspond to the submicroscopic level of thought although there is no explicit reference to “acid particles” there is an implicit suggestion that this solution is in some way different to a solution that does not produce a colour change.

Comparing Fig. 4 and 5 demonstrates that student talk in the domain of ideas occurs more often during the Litmus lesson (values shown in bold) specifically, there are 26 occurrences compared to zero during the Neutralisation lesson.

In Fig. 5, the teacher is recorded using symbolics during both the pre- and post-experimental phases of the Neutralisation lesson, whereas there is no evidence of students doing so during the experimental phase. In the Neutralisation lesson post-experimental phase excerpt (from Extract 6), the teacher provided a micro-scaffold for Student 1 to respond, “H plus”. This is likely to be indexical as there is no evidence to suggest that hydrogen ions nor the submicroscopic Arrhenius model of acids is understood.

Using the 3P framework to identify problematic communication

Using multiple levels of thought places an increased demand on the learners working memory which may impede learning (Johnstone, 1991). Comparing Fig. 4 and 5, only the Neutralisation lesson draws on all three levels of thought which may mean that the language used during this lesson places a greater demand on the learner's working memory than the language used during the Litmus lesson. As the students were not recorded working in the domain of ideas during the experimental phase of the Neutralisation lesson, this assumption seems probable. If increasing the demand on students’ working memory impedes learning, then the language used to present the Neutralisation lesson imposes a greater barrier to learning than the language used to present the Litmus lesson. It has been reported that students struggle to make the intended meaning when presented with multiple levels of thought, students may not make sense of them in the way the teacher intended (see for example, Lin and Chiu, 2010) thus, the potential for misunderstanding is also increased. The analysis using the 3P framework identifies three issues with the instructional communication used in the Neutralisation lesson.

Firstly, in the post-experimental phase excerpt (from Extract 6), the teacher repeatedly refers to “H plus”, as an oral abbreviation of hydrogen ions. In doing so, the teacher is implicitly moving between the submicroscopic and symbolic levels of thought. Instruction that moves between multiple levels of thought increases the extraneous cognitive load placed on the learner's working memory to the detriment of learning (Milenković et al., 2014).

Secondly, colour is a semiotic mode used to convey meaning (Pantaleo, 2012) the particle models drawn on the board (reproduced in Extracts 2 and 6) show the hydrogen ion in blue and hydroxide ion in blue and red. An expert will recognise that the hydrogen nucleus is common to both ions and be able to decode the teacher's meaning, but it may be distracting for a novice and add to the extrinsic cognitive load Miller et al. (2019).

Finally, in the pre-experimental phase excerpt (from Extract 2) the teacher states, “All the OHs have combined with all the Hs and made water…” and “Not acid, not alkali because all the OHs and Hs have joined to make water” in which the ellipsis of H⁺ to H and OH⁻ to OH has occurred. Ellipsis is defined as the deletion of linguistic elements that can be understood from contextual clues (Bussmann et al., 2006), here; the contextual clues are derived from the multimodal data available to the students. However, the resultant terms are incorrect and the unconscientious modelling of imprecise language compounds the difficulty in learning chemistry. If a learner's understanding of scientific language can be facilitated by combining the appropriate everyday language with scientific language (Rees et al., 2021) then resorting to the use of symbolics as an oral shorthand may be avoided.

The teacher's use of problematic language during the neutralisation lesson can be related to the commercially available course and assessment materials used by the school. These materials require 11 and 12 year-olds to represent neutralisation as the symbol equation for the formation of water from hydroxide ions and hydrogen ions (Gardom Hulme et al., 2013). This requirement disrupts curriculum coherence (Gardner et al. 2014) as the particle model and atoms are new ideas for the young learners, whereas ions and the Arrhenius model of acids are met in the 14–16 year-old's curriculum (UK Government, 2014). The ellipsis may have arisen because the teacher is avoiding introducing and discussing ions and charges, thus symbols and part-symbols are used without acknowledging the Arrhenius definitions underlying submicroscopic ideas. The ellipsis and the use of verbal shorthand indicate that both learning and teaching difficulties arise when knowledge is presented in steps (Danili and Reid, 2004) that disrupt the hierarchical sequence of scientific ideas (McPhail, 2021).

Outcomes from the application of the 3P Framework

Key words introduced by the teacher were tracked and counted throughout each phase of the lesson making the students’ up take of this language visible. The language used during the Litmus and Neutralisation lessons was compared using the 3P framework (Fig. 2) and the following observations were made:

• There were fewer incidences of learners using the language introduced by the teacher during the Neutralisation lesson

• None of the language associated with the domain of ideas was used by students in the experimental phase of the Neutralisation lesson

• Only the Neutralisation lesson used chemical symbols and drew on all three levels of thought

The analysis of the students’ language use during the Litmus lesson indicates that the learners have assimilated the language modelled by the teacher as both the key words and the pattern in which they were used were evident in the experimental phase.

RQ 2: To what extent can student talk be used to determine a practical task's effectiveness as a site for communicating chemistry?

The experimental phase for each lesson lasted approximately 40 minutes. Transcripts of student talk during the hands-on practical tasks were deductively coded into sequences corresponding to Mercer's (1995) typology of talk. Most of the student–student talk coded as cumulative; an example of which is provided in the excerpt below.

Litmus lesson experimental phase cumulative talk (from Extract 7)

Student 1: I’m doing the next one [row 44]

Student 2: You need to write it turns red [row 45]

Student 1: I have [row 46]

Student 1: No, I haven’t [row 47]

Student 2: You get the water thing and I’ll get litmus [row 48]

The student interaction is positive but there is a lack of discussion which results in a series of parallel statements rather than dialogue. Extract 7 and the cumulative talk moves described by the Talk ID Grid are available in Appendix 4.

There was only one instance of disputational talk identified from the transcripts, an example of which is provided in the excerpt below.

Litmus lesson experimental phase disputational talk (from Extract 8)

Student G Stop! [row 80]

Student G Write it down first [row 81]

Student L So blue no change [row 82]

Student G What you’ve already done it? [row 83]

Student G When we weren’t looking? [row 84]

The excerpt indicates competition within the group and individualised decision making rather than consensus. Extract 8 and the disputational talk moves described by the Talk ID Grid are available in Appendix 5.

One talk sequence from Group 1 during the Litmus task (Extract 3) was coded as exploratory talk as: the students were constructively critical of each other's ideas; and worked together to collect and analyse their data. This exploratory talk sequence was then deductively coded into the four quadrates of the Talk ID Grid which has been divided for clarity into Fig. 5, the domain of observables and Fig. 6, domain of ideas.

Fig. 6 Exploratory talk moves in the domain of observables in which: discursive moves characterise the manner in which students work together; and action moves which describe how students engage with each other's ideas.

The exploratory talk moves in domain of observables

The discursive moves in Fig. 5 indicate that the group has agreed that latecomer Student 3 can carry out the next test. The communication begins when Student 3 asks “Have we done this?” and establishes himself as a member of the group by using the inclusive pronoun “we”. The discourse moves in Fig. 5 at level 1 demonstrate that the group has entered dialogue, listening to, and responding to each other to reach the common goal of collecting results.

The level 2 exploratory talk action moves in the domain of observables reveal that Student 3 is both carrying out the practical task, and is actively considering the result's meaning by asking “So if they both don’t change then it has to be neutral?” Student 1 affirms and shares the data in his table of results to justify his response. These action moves indicate that the group are constructing knowledge and understanding of their actions in a process of active learning that may facilitate recall of the activity later.

The exploratory talk moves in domain of ideas

Fig. 7 level 1 content moves suggest that the students understand that testing with both red and blue litmus paper is necessary as Student 1 reports “No change, it is not an alkali” rather than suggesting it must be an acid. Furthermore, there is evidence of evaluation when Student 3 asks “Wait did any of the reds turn blue?” which Student 1 affirms and evidences his response by sharing his results table and names sodium hydroxide solution. The learners are seen to be focused on both collecting and interpreting results as they work. Analysis of the discussion reveals that the students are developing a shared understanding of the underlying concepts.

Fig. 7 Exploratory talk moves in the domain of ideas in which: content moves characterise students’ analysis of data; and purpose moves which describe how the students think together to understand their data.

The Level 2 purpose moves in the domain of ideas (Fig. 7) demonstrate how common knowledge has been constructed through the negotiation of the meaning of the experiment results. The exchange culminates when Student 3 states “Ok it's water, that makes sense.” aligning this new knowledge with his prior knowledge by recognising that deionised water, like tap water, must be neutral. Thus, Student 3 conveys his thinking through talk, and demonstrates that the analysis of the result has been internalised as, learning chemistry requires integration of the scientific viewpoint with existing ideas (Scott et al., 2011) suggesting that he will still be able to demonstrate this understanding later.

Outcomes from the application of the Talk ID Grid

Using the Talk ID Grid to identify and understand exploratory talk moves provides evidence of the students drawing on procedural and conceptual knowledge by working in both domains. The domain of observables is evidenced by students working collaboratively to collect the results required to complete the task. Whereas the students working in the domain of ideas is evidenced as the group build a shared understanding of the results.

Group 1 also used cumulative talk during the Litmus task, and it was the late arrival of Student 3 that triggered the change to exploratory talk. In Extract 3, Student 3 is showing interest and engagement with the Litmus task, but he is also accepted and supported by Student 1 and Student 2 and; it is this combination that facilitates collaboration and exploratory talk.

Fig. 7 the domain of ideas, documents Student 3 looking at his partner's results and asking, “So if they both don’t change it has to be neutral?” This initiates an extended talk sequence where the more knowledgeable partners help Student 3 construct new knowledge. From this perspective, the results table is performing a new role beyond dictating what is recorded to orchestrating dialogue.

Implications

The multimodal discourse analysis in this case study revealed that the two practical chemistry lessons observed provided three distinct linguistic opportunities. Firstly, during the pre-experimental phase the teacher introduced concepts and key words. Then, the student-centred experimental phase afforded learners the opportunity to use the teacher's language. Finally, the teacher-orchestrated post-experimental phase afforded deliberation and metalinguistic work. Developing language learning opportunities to support all students is an inclusive and equitable approach to facilitating learning in the laboratory.

The 3P framework could be used to plan practical lessons in which: key words and language patterns are foregrounded; and problematic language identified. Further, the lesson structure affords time for: student discussion during the practical task; and for discussion between the teacher and students when the hands-on activity is completed.

Patterns of language used by both the teacher and the students, replicated the structure of the results table. Understanding that a table of results may impact student talk, affords educators with the opportunity to decide in advance what they want the students to talk about and design the table accordingly. In this way, the teacher may regard the results table as a macro-scaffold for student talk (Nielsen and Hougaard, 2018).

The analysis of a sequence of exploratory talk using the Talk ID Grid, demonstrated the talk moves associated with an effective practical task. The Talk ID Grid operationalises talk moves as signposts of conceptual and procedural learning during practical work. This could help teachers to identify learning from student–student talk and to intervene when required with spontaneous micro-scaffolds such as repeating, revoicing, or questioning (Chin, 2006; Michaels and O’Connor, 2012, 2015; Nielsen and Hougaard, 2018).

Direct assessment is part of most teaching episodes: scanning the room to see if the task is complete; checking laboratory equipment is being used safely; listening to tones of voice; and noting body language in case someone needs help. Similarly, student talk moves described in the Talk ID Grid (available in the Appendix 1) could be used to indicate that a group needs help managing collaboration or discussion. As teachers become more familiar with the range of talk moves identified in the Talk ID Grid, the moves could be employed as signposts for assessment for learning (Black and Wiliam, 1998). For example, the Talk ID Grid could be used to classify group talk during a practical task to identify aspects of the intended learning that need to be reinforced in subsequent lessons. In addition, targeted modifications to the practical task could be devised to make future iterations more effective as a site for communicating chemistry.

Further work

Identifying the features of effective talk during a practical task provides an entry point for designing macro-scaffolds that facilitate the desired moves. The content and purposive moves associated with the domain of ideas in Fig. 7 were observed when students were completing the analysis column in the Litmus task table of results. Investigating the efficacy of the results table as a macro-scaffold offers scope for further work. The impact on student talk of including: 1, a column designed to initiate discussion of data such as, how one variable is affected by another, and 2, a column designed to stimulate reasoning about and discussion of the implication of their result could be investigated. Fig. 8 provides an example of how a results table designed as a macro-scaffold for student talk during the Neutralisation task may be constructed.

Fig. 8 An example of a possible Neutralisation task results table designed to be a student talk macro-scaffold.

The discursive and action moves associated with the domain of observables shown in Fig. 6, relate to the ways the students are interacting whilst carrying out the practical task. During this episode of exploratory talk, the students were working collaboratively (Kirschner et al., 2009) to create common knowledge. Adopting protocols such as Lab Roles and Lab Talk, as described in our earlier work (Hennah et al., 2022), could facilitate more episodes of collaboration and exploratory talk during future iterations.

Research has shown that formal and explicit instruction in collaborative skills is requisite for classroom collaboration to occur (Le et al., 2018). The co-occurrence of collaboration and exploratory talk observed in this case study is of interest, as approaches known to facilitate collaboration may also stimulate exploratory talk.

Conclusion

This work has demonstrated mechanisms by which the laboratory may be recast as a site for teaching and learning the language of chemistry. Using the 3P framework to plan and deliver practical lessons would ensure that the time and opportunity are provided for learners to discuss and build an understanding of the task and its underlying chemical concepts. The resultant student talk may in turn be evaluated using the Talk ID Grid. The relationship between academic success, vocabulary, literacy skills, and social mobility has long been apparent (Hart and Risley, 2003) which, in conjunction with changing population profiles, suggests the need for a greater focus on language in chemistry education. It is hoped that this work will provide colleagues with tools and support to help do so in the laboratory.

Limitations

As a small-scale case study, the generalisability of any conclusions is very limited due to the specificity of the case. Further, the qualitative methods employed here are subjective; transcription is a transduction as semiotic material is moved from one mode to another and reflects both the research aims and directs the research findings. Although, data was cleaned by the researcher repeatedly watching the recordings and comparing the transcripts with field notes and the learners written work, validation by a second experienced researcher would further secure the outcomes. Furthermore, unconscious bias must also be acknowledged, particularly as the researcher also teaches chemistry in the same school.

The study and the frameworks presented were designed by a teacher who, motivated by an apparent language barrier in teaching and learning chemistry, sought to understand language, and learning in practical lessons. In doing so, future interventions could be targeted to better support learning in the context of the school chemistry laboratory observed, however, the study may also be useful to other educators, and researchers seeking to foster the quality of student talk and collaboration.

Conflicts of interest

There are no conflicts to declare.

Appendix 1: The Talk ID Grid identifies and describes talk moves used to code student talk

The Talk ID Grid to for the Domain of observables.

The Talk ID Grid to for the domain of ideas.

Appendix 2: learning objectives and practical task methods

Litmus lesson Learning Objective: Categorise substances as acid, alkali, or neutral using experimental observations.

Litmus task method

1. Tear each piece of litmus paper into three smaller pieces.

2. Place a small piece of red litmus paper into one well of the spotting tile.

3. Using a pipette, add a drop of sulphuric acid to the red litmus paper.

4. Record your observation in the results table (shown below).

5. Repeat steps 1 to 4 with a small piece of blue litmus paper.

6. Using the two litmus paper results complete the analysis column of the table.

7. Repeat steps 1 to 6 with the remaining five solutions.

8. Dispose of the pieces of litmus paper in the waste bin.

Neutralisation lesson Learning Objective: Describe how pH changes in neutralisation reactions and relate these changes to the colour of universal indicator.

Neutralisation task method

1. Pour 10 cm³ hydrochloric acid into a conical flask.

2. Add a few drops of universal indicator and swirl the beaker carefully.

3. Pour 9 cm³ sodium hydroxide solution into a second beaker.

4. Using a pipette, carefully add the sodium hydroxide drop by drop to the conical flask containing acid and universal indicator. Keep swirling.

5. As the sodium hydroxide is added note the colour changes and the number of drops added to produce the change.

Appendix 3: Exemplar extracts of multimodal communication data from the practical lessons

Extract 1: Litmus lesson pre-experimental phase IRF sequence

Extract 2: Neutralisation lesson pre-experimental phase, teacher monologue

Extract 3: Litmus lesson experimental phase, Group 1 exploratory talk

Extract 4: Neutralisation lesson experimental phase, Group 2 Cumulative Talk

Extract 5: Litmus lesson post-experimental phase IRF sequence

Extract 6: Neutralisation lesson post-experimental phase IRF sequence

Appendix 4: An extract of cumulative talk deductively coded into Talk ID Grid

Extract 7: Experimental phase Litmus task Group 1 Cumulative Talk

Cumulative talk moves in the domain of observables in which: discursive moves characterise the manner in which students work together; and action moves which describe how students engage with each other's ideas.

Cumulative talk moves in the domain of ideas in which: content moves characterise students’ analysis of data; and purpose moves which describe how the students think together to understand their data.

Appendix 5: The extract of disputational talk deductively coded into Talk ID Grid

Extract 8: Experimental phase Litmus Task Group 2 Disputational Talk

Disputational talk moves in the domain of observables in which: discursive moves characterise the manner in which students work together; and action moves which describe how students engage with each other's ideas.

Disputational talk moves in the domain of ideas in which: content moves characterise students’ analysis of data; and purpose moves which describe how the students think together to understand their data.

References

1. Abrahams I. and Millar R., (2008), Does practical work work? A study of the effectiveness of practical work as a teaching and learning method in school science, Int. J. Sci. Educ., 30(14), 1945–1969.
2. Abrahams I. and Reiss M. J., (2012), Practical work: Its effectiveness in primary and secondary schools in England, J. Res. Sci. Teach., 49(8), 1035–1055.
3. Andersson J. and Enghag M., (2017), The relation between students’ communicative moves during laboratory work in physics and outcomes of their actions. Int. J. Sci. Educ., 39(2), 158–180.
4. Ary D., Jacobs L. C., Irvine C. K. S. and Walker D., (2018), Analysing and reporting qualitative results, Introduction to Research in Education, Belmont: Cengage Learning, pp. 480–511.
5. Barnes D., (2010), Why talk is important, English Teach.: Practice Critique, 9(2), 7–10.
6. Baxter P. and Jack S., (2008), Qualitative case study methodology: study design and implementation for novice researchers, Qualitative Rep., 13(4), 544–559.
7. Bezemer J. and Mavers D., (2011), Multimodal transcription as academic practice: A social semiotic perspective, Int. J. Soc. Res. Methodol., 14(3), 191–206.
8. Black P. and Wiliam D., (1998), Assessment and classroom learning, Assess. Educ.: Prin., Policy Practice, 5(1), 7–74.
9. Bree R. T. and Gallagher G., (2016), Using Microsoft Excel to code and thematically analyse qualitative data: a simple, cost-effective approach, All Ireland Journal of Higher Education, 8(2), https://ojs.aishe.org/index.php/aishe-j/article/view/281.
10. British Association for Applied Linguistics, (2021), Recommendations on Good Practice in Applied Linguistics [Online] Available at https://www.baal.org.uk/wp-content/uploads/2021/03/BAAL-Good-Practice-Guidelines-2021.pdf (Accessed 14/6/2022).
11. British Educational Research Association, (2018), Ethical guidelines for educational research, 4th edn, Available at: https://www.bera.ac.uk/researchers-resources/publications/ethical-guidelines-for-educational-research-2018 (Accessed 17/11/2022).
12. Bullock A., (1975), The Bullock Report. A language for life. Education, 145, i–viii. [Online] Available at http://www.educationengland.org.uk/documents/bullock/bullock1975.html (Accessed 5/8/2022).
13. Bussmann H., Kazzazi K. and Trauth G., (2006), Routledge dictionary of language and linguistics, Routledge.
14. Byrne M., Johnstone A. H. and Pope A., (1994), Reasoning in science: a language problem revealed? School Sci. Rev., 75, 103–103.
15. Cassels J. R. T. and Johnstone A. H., (1984), The effect of language on student performance on multiple choice tests in chemistry, J. Chem. Educ., 61(7), 613.
16. Chin C., (2006), Classroom interaction in science: teacher questioning and feedback to students’ responses, Int. J. Sci. Educ., 28(11), 1315–1346.
17. Coulthard M. and Condlin C. N., (2014), An introduction to discourse analysis, Routledge.
18. Danili E. and Reid N., (2004), Some strategies to improve performance in school chemistry, based on two cognitive factors, Res. Sci. Technol. Educ., 22(2), 203–226.
19. Dalton-Puffer C., (2011) Content-and-language integrated learning: From practice to principles? Ann. Rev. Appl. Linguistics, 31, 182.
20. Dalton-Puffer C., Nikula T. and Smit U., (ed.), (2010), Language use and language learning in CLIL classrooms, John Benjamins Publishing, vol. 7.
21. de Oliveira L. C., (2016), A language-based approach to content instruction (LACI) for English language learners: Examples from two elementary teachers, Int. Multiling. Res. J., 10(3), 217–231.
22. Flewitt R., (2011), Bringing ethnography to a multimodal investigation of early literacy in a digital age, Qual. Res., 11(3), 293–310.
23. Gardner A. L., Bybee R. W., Enshan L. and Taylor J. A., (2014), Analyzing the coherence of science curriculum materials, Curric. Teach. Dialogue, 16(1/2), 65–86.
24. Gardom Hulme P., Locke J., Reynolds H. and Grevatt A., (2013), Kerboodle: Activate 1 Student Book, Activate 1 Lessons, Resources and Assessment Retrieved 3rd Dec, 2021 https://www.kerboodle.com/.
25. Gatsby (2017), Good Practical Science Report. Retrieved 3rd Dec, 2021, from https://www.gatsby.org.uk/uploads/education/reports/pdf/good-practical-science-report.pdf.
26. Gee J. P., (2008), A sociocultural perspective on opportunity to learn, Assessment, Equity, Opportunity Learn, 76–108.
27. Georgiadou A. and Tsaparlis G., (2000), Chemistry teaching in lower secondary school with methods based on: (a) psychological theories; (b) the macro, representational, and submicro levels of chemistry, Chem. Educ. Res. Practice, 1(2), 217–226.
28. Gilbert J. K., (2010), The role of visual representations in the learning and teaching of science: an introduction, Asia-Pac. Forum Sci. Learn. Teach., 11(1), 1–19.
29. Halliday M. A. K., (2004), ‘Three aspects of children's language development: learning language, learning through language, learning about language’, Webster, J., (ed.) in The Language of Early Childhood, London: Continuum, pp. 308–326.
30. Hart B. and Risley T. R., (2003), The early catastrophe: the 30 million word gap by age 3, Am. Educ., 27(1), 4–9.
31. Hennah N., Newton S. and Seery M. K., (2022), A holistic framework for developing purposeful practical work, Chem. Educ. Res. Pract., 23, 582–598.
32. Hofstein A., (2017), The role of laboratory in science teaching and learning, Science Education, Brill, pp. 355–368.
33. Jiménez-Aleixandre M. P. and Reigosa C., (2006), Contextualizing practices across epistemic levels in the chemistry laboratory, Sci. Educ., 90(4), 707–733.
34. Johnstone A. H., (1982), Macro- and micro-chemistry, School Sci. Rev., 64(227), 377–379.
35. Johnstone A. H., (1991), Why is science difficult to learn? Things are seldom what they seem, J. Comput. Assis. Learn., 7(2), 75–83.
36. Johnstone A. H. and Wham A. J. B., (1979), A Model for Undergraduate Practical Work, Educ. Chem., 16(1), 16–17.
37. John-Steiner V. and Mahn H., (1996), Sociocultural approaches to learning and development: a Vygotskian framework, Educ. Psychologist, 31(3–4), 191–206.
38. Kirschner F., Paas F. and Kirschner P. A., (2009), A cognitive load approach to collaborative learning: united brains for complex tasks, Educ. Psychol. Rev., 21(1), 31–42.
39. Le H., Janssen J. and Wubbels T., (2018), Collaborative learning practices: teacher and student perceived obstacles to effective student collaboration, Cambridge J. Educ., 48(1), 103–122.
40. Lemke J. L., (1982), Classroom Communication of Science, Final Report.
41. Lemke J. L., (1990), Talking science: Language, learning, and values, Norwood: Ablex Publishing Corporation.
42. Lin J. W. and Chiu M. H., (2010), The mismatch between students’ mental models of acids/bases and their sources and their teacher's anticipations thereof, Int. J. Sci. Educ., 32(12), 1617–1646.
43. Lunetta V. N., Hofstein A. and Clough M. P., (2007), Learning and teaching in the school science laboratory: an analysis of research, theory, and practice, Handbook Res. Sci. Educ., 2, 393–441.
44. Markic S., (2015), Chemistry teachers’ attitudes and needs when dealing with linguistic heterogeneity in the classroom, Affective dimensions in chemistry education, Berlin, Heidelberg: Springer, pp. 279–295.
45. Markic S. and Childs P. E., (2016), Language and the teaching and learning of chemistry, Chem. Educ. Res. Practice, 17(3), 434–438.
46. McPhail G., (2021), The search for deep learning: a curriculum coherence model, J. Curriculum Studies, 53(4), 420–434.
47. Mercer N., (1995), The guided construction of knowledge: Talk amongst teachers and learners, Multilingual matters.
48. Mercer N., (2002), Words and minds: How we use language to think together, Routledge.
49. Michaels S. and O’Connor C., (2012), Talk science primer, Cambridge, MA: TERC. http://searkscience.pbworks.com/w/file/fetch/67803311/18-TalkScience_PrimerArticle.pdf.
50. Michaels S. and O’Connor C., (2015), Conceptualizing talk moves as tools: professional development approaches for academically productive discussion, Socializing Intelligence Through Talk Dialogue, 347, 362.
51. Milenković D., Segedinac M., Hrin T. and Cvjetićanin S., (2014), Cognitive load at different levels of chemistry representations, Hrvatski časopis za odgoj i obrazovanje, 16(3), 699–722.
52. Millar R., (2013), Improving science education: Why assessment matters, Valuing assessment in science education: Pedagogy, curriculum, policy, Dordrecht: Springer, pp. 55–68.
53. Millar R. and Abrahams I., (2009), Practical work: making it more effective, School Sci. Rev., 91(334), 59–64.
54. Miller P., Hazan-Liran B. and Cohen D., (2019), Does task-irrelevant colour information create extraneous cognitive load? Evidence from a learning task, Quart. J. Exp. Psychol., 72(5), 1155–1163.
55. National Academies of Sciences, Engineering, and Medicine (2018), English Learners in STEM Subjects: Transforming Classrooms, Schools, and Lives, Washington, DC: The National Academies Press. https://doi.org/10.17226/25182.
56. Nielsen B. L. and Hougaard R. F., (2018), Scaffolding students' reflective dialogues in the chemistry lab: challenging the cookbook, Research, Practice and Collaboration in Science Education Proceedings of the ESERA 2017 Conference: Proceedings of ESERA 2017, Dublin City University, pp. 2237–2246.
57. Nikula T., (2015), Hands-on tasks in CLIL science classrooms as sites for subject-specific language use and learning, System, 54, 14–27.
58. O’Halloran K. L., (2005), Mathematical discourse: Language, symbolism and visual images, London/New York: Continuum.
59. O’Halloran K. L., (2015), The language of learning mathematics: a multimodal perspective, J. Mathematical Behavior, 40, 63–74.
60. Pantaleo S., (2012), Meaning-making with colour in multimodal texts: an 11 year-old student's purposeful ‘doing’, Literacy, 46(3), 147–155.
61. Quílez J., (2021), Supporting Spanish 11th grade students to make scientific writing when learning chemistry in English: the case of logical connectives, Int. J. Sci. Educ., 43(9), 1459–1482.
62. Rees S. W., Kind V. and Newton D., (2018), Can language focused activities improve understanding of chemical language in non-traditional students? Chem. Educ. Res. Pract., 19(3), 755–766.
63. Rees S., Kind V. and Newton D., (2021), The development of chemical language usage by “non-traditional” students: the interlanguage analogy, Res. Sci. Educ., 51(2), 419–438.
64. Reiss M. J. and Abrahams I., (2015), The assessment of practical skills, Sch. Sci. Rev., 357, 40–44.
65. Russell C. B. and Weaver G. C., (2011), A comparative study of traditional, inquiry-based, and research-based laboratory curricula: impacts on understanding of the nature of science, Chem. Educ. Res. Pract., 12(1), 57–67.
66. Sacks H., Schegloff E. A. and Jefferson G., (1978), A simplest systematics for the organization of turn taking for conversation, Studies in the organization of conversational interaction, Academic Press, pp. 7–55.
67. Sandi-Urena S., Cooper M. M., Gatlin T. A. and Bhattacharyya G., (2011), Students' experience in a general chemistry cooperative problem based laboratory, Chem. Educ. Res. Pract., 12(4), 434–442.
68. Scott P., Mortimer E. and Ametller J., (2011), Pedagogical link-making: a fundamental aspect of teaching and learning scientific conceptual knowledge, Studies Sci. Educ., 47(1), 3–36.
69. Silliman E. R., Wilkinson L. C. and Brea-Spahn M., (2018), Writing the science register and multiple levels of language, Language, literacy, and learning in the STEM Disciplines: How language counts for English learners, 115–139.
70. Sinclair J. M. and Coulthard R. M., (1975), Towards an Analysis of Discourse: The English Used by Teachers and Pupils, London: Oxford University Press.
71. Stake R. E., (1995), The art of case study research, Sage.
72. Sweller J., van Merriënboer J. J. and Paas F., (2019), Cognitive architecture and instructional design: 20 years later, Educ. Psychol. Rev., 31(2), 261–292.
73. Talanquer V., (2011), Macro, submicro, and symbolic: the many faces of the chemistry “triplet”, Int. J. Sci. Educ., 33(2), 179–195.
74. Tiberghien A., (2000), Designing teaching situations in the secondary school, R. Millar, J. Leach and J. Osborne (ed.) in Improving science education: The contribution of research, Buckingham: Open University Press, pp. 27–47.
75. Tsaparlis G., Kolioulis D. and Pappa E., (2010), Lower-secondary introductory chemistry course: a novel approach based on science-education theories, with emphasis on the macroscopic approach, and the delayed meaningful teaching of the concepts of molecule and atom, Chem. Educ. Res. Pract., 11(2), 107–117.
76. Tobin K., (1990), Research on science laboratory activities: in pursuit of better questions and answers to improve learning, School Sci. Mathematics, 90(5), 403–418.
77. Toulmin S., (1972), Conceptual change and the problem of relativity, Critical Essays Philosophy of RG Collingwood, 201–221.
78. UK Government, (2014), Science programmes of study: key stage 4 National curriculum in England (retrieved 8/4/2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/381380/Science_KS4_PoS_7_November_2014.pdf.
79. Vygotsky L. S., (1987), The Collected Works of L.S. Vygotsky, Thinking and Speaking, New York: Plenum Press, vol. 1.
80. Vygotsky L. S., (1978), Mind in Society: The Development of Higher Psychological Processes, Cambridge, MA: Harvard University Press.
81. Yin R. K., (2003), Case study research: Design and methods, 3rd edn, Thousand Oaks, CA: Sage.
82. Yore L. D. and Treagust D. F., (2006), Current realities and future possibilities: language and science literacy—empowering research and informing instruction. Int. J. Sci. Educ., 28, 291–314.

---