CC BY-ND
11
Building Scientific Knowledge in English: Integrating Content, Cognition and Communication in Secondary School CLIL Biology
Pilar Gerns 12
' University of Navarra, Pamplona, Spain 2 UNIR, Rioja, Spain
ABSTRACT
Background: The focus of this paper is on Dalton-Puffer's construct of the Cognitive Discourse Function (CDF) (2013), which offers CLIL teachers a practical framework through which they can more easily understand the complex idea of integrating the content, cognition, and language required for their subject. These functions have mainly been addressed from classroom observations or task prompts, and little is known about their teachability and effectiveness on students' content knowledge.
Purpose: This paper explores whether the CDF of 'comparing' (a subcategory of 'classify') can be taught to Spanish seventh-grade CLIL biology students (N = 37) and examines the effect of teaching it explicitly on their written performance.
Method: An operational framework was developed to define this CDF and an exploratory study was performed in which students were asked to hand in written comparisons. Quantitative and qualitative pre-and post-tests were applied.
Results: Significant results were obtained for the experimental groups, which improved in both content and language learning, scoring higher on inclusion of content points, justification of their scientific claims, concept formation and use of lexico-grammatical forms. Conclusion: These findings add to our understanding of the importance of integrating cognition and language in teaching and learning natural sciences, within which CDFs can be a useful starting point.
KEYWORDS
Content and Language Integrated Learning (CLIL), Cognitive Discourse Function (CDF), secondary school science education, knowledge construction
https://doi.org/10.17323/jie.2023.17569
Citation: Gerns P. (2023). Building Scientific Knowledge in English: Integrating Content, Cognition
INTRODUCTION
and Communication in Secondary School CLIL Biology. Journal of Language and Education, 9(3), 52-78. https://doi.org/10.17323/jle.2023.17569
Correspondence:
Pilar Gerns, [email protected]
Received: July 03, 2023 Accepted: September 15, 2023 Published: September 30, 2023
Across Europe and many other parts of the world where English is not the vehicular language, school subjects are now being taught in English. In European countries, this approach is generally understood within the paradigm of Content and Language Integrated Learning (CLIL): when suitable didactic techniques are used, students can learn scientific content at the same time as they improve their English language skills and begin to develop scientific literacy in what is effectively the world language of science. However, the actual process of integrating content and lan-
guage in CLIL classrooms still remains a challenge (Breeze et al., 2014; Llinares, 2015; Villabona & Cenoz, 2021), not least because CLIL teachers seem to continue giving their classes relying on their previous, traditional didactics, where content and language are taught separately. This suggests that the idea of integrated learning has not yet fully reached the classroom. There is a need to bring existing research into CLIL classroom practice (Llinares et. al., 2012, 2015; Nikula et al., 2016) and provide CLIL teachers with concrete, pedagogical tools they can use to foster integrated learning, that is, co-development of the cognitive, content-oriented demands of the subject
and the language that students need to express their emergent knowledge effectively (Nikula, 2017; Morton, 2020).
Various European research initiatives address the question of integrating content and language learning in classroom practice. These include the Council of Europe subject language descriptors (Vollmer, 2011), the genre-based approach (Llinares et al., 2012; Lorenzo, 2017) and the Graz-group Pluriliteracy Teaching for Learning (PTL) model (Meyer et al., 2015; Coyle & Meyer, 2021; Coyle et al., 2023). One promising departure in this area is the construct of the Cognitive Discourse Function (CDF), proposed by Dalton-Puffer (2013), which offers CLIL teachers a practical conceptual framework they can use in their CLIL planning, teaching, and assessment (Morton, 2020).
The novelty of this approach lies in its interdisciplinary nature, which offers content and language teachers a shared pedagogical tool that unites contributions from both disciplines, namely the notions of thinking skills in science and other subject education and the notion of discourse function in applied linguistics (Dalton-Puffer, 2013; Bauer-Marschallinger, 2022). Dalton-Puffer essentially identifies a 'zone of convergence' between content and language learning objectives, recognising that learning in school tends to occur through a limited set of cognitive operations (such as 'explain' or 'define'), which are expressed using a finite set of verbs. The cognitive demands of content courses can thus be linked to specific linguistic patterns, which allow to them be identified and taught (Dalton-Puffer, 2013). The CDF can thus be considered an 'entry point' for students that introduces them to subject-specific ways of thinking and communicating, which are an integral component of subject literacies (Veel, 1997; Rose & Martin, 2012; Zwiers, 2014; Polias, 2015).
Table 1
Cognitive Discourse Functions
In one landmark paper, Dalton-Puffer (2013) presents an operational framework of the most important CDFs in and for school learning, bringing together 50 different functions from different curricula and learning frameworks, which are then systematized according to their communicative intentions yielding seven main types, as shown in Table 1. These are: categorise, define, describe, evaluate, explain, explore, and report.
The idea is basically that when students are asked to perform a CDF-task, such as to 'explain' the process of photosynthesis, they have to process the subject contents cognitively and verbalize this process through some corresponding linguistic forms. If properly scaffolded, the effort to express the newly learned knowledge in English will foster students' subject learning, leading them over time to more intertwined, deeper learning (Vollmer, 2011; Meyer et al., 2015, 2018; Morton, 2020).
Research on CDFs so far has mainly been conducted from descriptive classroom observations or by eliciting students' CDF use through concrete tasks, analysing whether and how CLIL users acquire these discursive forms without explicitly teaching them (see Lose, 2007; Breeze & Dafouz, 2017; Lorenzo, 2017; Dalton-Puffer et al., 2018; Evnitskaya & Dalton-Puffer, 2020; Doiz & Lasagabaster, 2021; Llinares & Nashaat-Sobhy, 2021; Salvador-Garcia & Chiva-Bartoll, 2022; Whittaker & McCabe, 2023; Llinares & Nikula, 2023). The results of these studies show that CDFs do occur in the course of ordinary CLIL classes, regardless of the school subjects and students' age. However, they also pinpoint the fact that almost no explicit CDF teaching and learning takes place and that, in general, little attention is paid to the functional side of subject-specific language in CLIL contexts. The data present low incidences of explicit metatalk and few output opportunities for students, who thus fail to develop a pro-
Type
Members
Communicative Intention
categorise categorise, classify, compare, contrast, exemplify, match, structure, subsume
define define, identify, characterise
describe describe, label, identify, name, specify
evaluate evaluate, argue, judge, take a stance, critique, comment, reflect, judge
explain explain, reason, express cause/effect
explore explore, hypothesise, predict, speculate, guess, estimate, simulate
report report, inform, summarise, recount, narrate, present, relate
I tell you how we can cut up the world according to certain ideas.
I tell you about the extension of this object of specialised knowledge.
I tell you details of what I can see (also metaphorically). I tell you what my position is vis a vis X.
I tell you about the causes or motives of X. I tell you something that is potential (i.e., non-factual).
I tell you something external to our immediate context on which I have a legitimate knowledge claim.
Note. From "Cognitive Discourse Functions in Austrian CLIL Lessons," by C. Dalton-Puffer, S. Bauer-Marschallinger, K. Brückl, V. Hofmann, J. Hopf, L. Kröss, & L. Lechner, 2018, European Journal of Applied Linguistics, 6(1), pp. 5-29 (https://doi.org/10.1515/eujal-2017-0028). Copyright 2018 by the De Gruer.
ductive command of CDFs. As natural acquisition of these academic functions seems not to occur automatically when teaching CLIL, research has stressed the need for more explicit teaching and practice of CDF forms (Dalton-Puffer, 2007; 2017; Meyer et al., 2018; Morton, 2020; Nashaat- Sob-hy, 2020; Coyle et al., 2023).
Some first steps have already been taken in this direction by working on the conceptual and linguistic framework of the different CDFs using predominantly a Systemic Functional Linguistics (SFL) approach to make them operable. Such is the case for 'categorize' (Evnitskaya & Dalton-Puffer, 2020), 'define' (Nashaat- Sobhy, 2020; Llinares & Nashaat- Sobhy, 2021), 'describe' (Dalton-Puffer, 2004), 'evaluate' (Llinares & Dalton-Puffer, 2015; Whittaker & McCabe, 2023; Hasen-berger, ongoing), 'explain' (Dalton-Puffer, 2004; Lose, 2007; Connolly, 2019), and 'predict' (a subcategory of 'explore') (Dalton-Puffer, 2007), though only a few have been validated in terms of usefulness by developing explicit teaching programmes. Work in this direction includes a study by Breeze and Gerns (2019) in a secondary-level CLIL history class, where some general CDF-related features were explicitly taught for 'describing' and 'explaining', four doctoral theses (Connolly, 2019; Gerns, 2021; Bauer-Marschallinger, 2022; Hasenberger, ongoing) and Nashaat-Sobhy's (2020) validated framework for working with the CDF 'defining' in the university setting. Classroom materials are now being developed to teach, scaffold and assess these CDFs in the CLIL classroom1 (see Coetzee-Lachmann, 2019; DeBoer & Leontjev, 2020; Del Pozo & Llinares, 2021; Coyle et al., 2023). However, more research is clearly needed to clarify the cognitive and linguistic parameters of the various CDFs across school subjects and test their practical applicability, in particular, in secondary L2 education, where the students' content, cognitive and language skills are undergoing intense development.
This study aims to contribute to this ongoing research by first defining the CDF of 'comparing' (a subcategory of 'categorise') in the context of science, where it constitutes a fundamental method and offers a good starting point to introduce students to scientific working culture and its specific modes of thinking and communicating, which are a key component of scientific literacy (Lemke, 1990; Roberts & Bybee, 2014). By enabling students to articulate their own scientific knowledge in a much clearer, more organised and professional way, teachers thus empower them to develop the necessary cognitive and linguistic skills to take part in society's wider scientific debates in and beyond school (Mercer, 1995; Norris & Phillips, 2009; Oliveira & Weinburgh, 2017; Coyle & Meyer, 2021).
Second, this paper aims to examine the effect of explicitly teaching the CDF of 'comparing' on students' subject-based written performance in a secondary school (7 grade) CLIL science class. To this end, some quantitative and qualitative pre-and post-test results will be shown from an empirical study in which the CDF of 'comparing' was explicitly taught to a group of Spanish CLIL biology students.
Defining the CDF of 'Comparing' in CLIL
To determine whether explicit teaching of the CDF of comparing promotes integrated learning between contents and language, an interdisciplinary approach was adopted, which integrates contributions made by the natural, educational, and linguistic sciences. This is briefly presented below.
'Comparing' in Natural Sciences
The act of comparing constitutes one of our first and most natural forms of human thought. From childhood onwards, we interact with the world by making comparisons, grouping things into pairs, and examining their shared and unshared features, that is, their similarities and differences. These comparison-based discoveries are nothing other than our epistemic way of making the world understandable and operational.
However, comparing does not only present a natural human way of thinking - it is also a fundamental method of scientific inquiry, integral to our modern sciences (Darian, 2003; Martinez, 2018). The idea of studying the living word through a comparative approach dates back to the Ancient Greeks, when Aristotle was one of the first to analyse the physiology of dissected animals and plants by identifying patterns of similarity and difference (Carpi & Egger, 2008). This method has developed since and constitutes an essential, distinctive methodology for biological science, being used to classify plants and animals into taxonomies, or to investigate comparative anatomy in the study of evolution, heredity, and adaptation to environmental pressures (Flan-nery, 2010). It is thus a core cognitive skill in biology, and an essential prerequisite for disciplinary literacy. In fact, comparing is frequently used in biology classes, for example, when viruses are taught together with bacteria, or the concept of warm-blooded with that of cold-blooded. Teaching students how to compare can be a good starting point to introduce them to comparative methods and help them think like scientists.
'Comparing' in Education
In education and cognitive psychology, comparing is understood to be an important component in terms of lower-or-
LongAd-CLIL project (2019-2021). A longitudinal corpus-based analysis of advancedness in CLIL. https://uam-CLIL.org/longad-CLIL/
der thinking skills (Bloom, 1956). It is defined as one of the most basic analytical lower-order thinking skills, grouped together with others (such as 'matching', 'classifying', 'error analysis', 'generalizing' and 'specifying'), which are essential to carry out more advanced and complex operations (such as 'define', 'predict', or 'classify'). Furthermore, different taxonomies (see Anderson & Krathwohl, 2001; Marzano, 2001; Biggs & Tang, 2011) situate comparing at the crossroads between the cognitive dimensions of 'understand' and 'analyse' and the knowledge dimension of concepts. The reason is that comparing consists of identifying commonalities between at least two items, which often requires moving from a surface, experiential-driven understanding towards a more abstract and deconstructive one.
Studies have shown that teaching how to compare properly in content areas seems to have positive effects on students' learning (Hammann & Stevens, 2003; Silver, 2007; Goldstone, 2010; MacArthur & Philippakos, 2010; Clark et al., 2020). For example, it helps to develop: (1) a structured and orderly way of thinking, (2) abstract and relational ways of learning, (3) detail-focused, and longer-lasting comprehension, since two things taught together seem to prove more memorable than one, and ultimately (4) advanced literacy skills.
'Comparing' in Linguistics
According to the Oxford English Dictionary (OED) (1989), the verb 'compare' is a late Middle English word, derived from the old French word 'comparer' and the Latin 'compara-re' ('com'- with + 'par'- equal; 'compar'-like), meaning literate to 'pair' or 'match together'. The transitive verb 'to compare' and its corresponding noun form 'comparison' can be defined in either one of the following ways:
(1) 'to speak of or represent as similar'; 'to make like'; 'to liken'.
Table 2
Five-Point Framework for the CDF of 'Comparing'
(2) 'to place together so as to point out the similarities and differences of (two or more things)'.
Based on these definitions, 'comparing' seems to be characterized by 'matching' or 'bringing' at least two relatable items together, that is, 'likening' them, and analysing them on some shared and unshared features, i.e., on their similarities and differences.
Academic discourse functions have been studied in the context of L2 teaching, in particular, by researchers from English for Specific Purposes (ESP) and Halliday's Systemic Functional Linguistics (SFL) tradition (Halliday & Matthiessen, 2004), which address the importance of such functions in academic writing (Polias, 2015; Smith, 2019). 'Compare' is considered one of the most common discourse functions in expository text, alongside others such as sequencing, classification, explanation, cause-effect, and problem-solution (De La Paz & McCutchen, 2010). This can also be regarded as an aspect of Cognitive Academic Language Proficiency (CALP) (Cummins, 2008), which is crucial for students' academic development (Llinares et al., 2012; Morton, 2020).
It is clear that cognitive functions (such as 'comparing') are tied to specific lexico-grammatical structures (such as 'is like', 'different to' or 'unlike') which allow teaching and learning through them. Considerable research has been conducted to clarify the analytical patterns of these discourse functions (see Cheong, 1978; Widdowson, 1979; Darian, 2003; Dixon; 2005; Huddleston, 2017; Evnitskaya & Dalton-Puffer, 2020).
Five-Point Framework for 'Comparing'
Based on the previous literature, I developed an operational framework that brings these contributions together. As a result, the act of comparing can be defined through at least five elements, shown in Table 2, each of which can be triggered by a corresponding question. This framework relies, in particular, on the work of Raphael and Kirschner (1985),
Components
Questions
Items being compared: "topic" (X) + "target" (Y) Points of comparison: "criteria"
Types of comparisons: general, specific comparison; similarities, differences.
What two items are compared? What is the "topic" and what is the "target"?
What (points) can these items be compared on? Can you name the criteria explicitly? Can you use the same criteria for both groups?
What type of comparison will be performed? Are the items being compared 1. in a general/ specific way or 2. in terms of similarities and differences. Are these points reasoned?
Language of and for comparing: "comparitor"
How do we express a comparative relationship? What lexico-grammatical forms are used?
Comparative structure: parallel organization
How do we structure the comparative points? In a block-method, or a point-by-point one?
Widdowson (1979), Cheong (1978), Darian (2003), Dixon (2005), Polias (2015) and Huddleston (2017), as they present some concrete analytical descriptions of how to make a scientific comparison.
The first element addresses the question of what two items will be compared, referred to as 'topic' and 'target' (Cheong, 1978). The second element consists of establishing the criteria on which these items can be compared on, which are variously known as the 'parameter' or 'basis' of a comparison (Widdowson, 1979; Dixon, 2005). These points are usually abstract concepts. Then, the third element refers to the conceptual categories a comparison can adopt, which include Halliday and Hasan's (1976) general and specific comparison and a comparison in terms of similarities, differences and matters of degree (Cheong, 1978; Darian, 2003; Halliday & Matthiessen, 2014; Polias, 2015). Fourth is the language of and for comparing, which are the specific lex-ico-grammatical forms and structures (such as 'similar to', 'different') through which comparative relationships can be expressed, known as the comparitor. Significant research has been done in ESP and SFL to clarify these forms. Finally, a comparison is also characterised by its parallel structure, according to which items can be compared and contrasted in a systematic way. See Gerns (2023) for further analysis of these aspects.
Exploratory Classroom Study
Rationale and Research Questions
Given the importance of validating explicit CDF-based teaching tools and verifying their effect not only on students' language but also on their content and cognitive learning (Badertscher & Bieri, 2009; Bauer-Marschallinger, 2022), the present paper aims to validate the CDF framework for comparing and establish whether explicitly teaching it in a CLIL science context can have a positive effect on students' subject knowledge. To do this, a classroom-based study was designed and executed with secondary CLIL biology students (7th grade, age 12-13), who were previously divided into a study and a control group. They were asked to hand in three written comparisons, which served as pre-and post-tests. The idea was to compare the results obtained before and after the instruction to examine whether the study group students displayed their content knowledge significantly better after the instruction phase than the control group.
The following research questions were addressed:
Can explicit CDF-based teaching (CDF'compare) have a significant effect on students' written subject performance, considering:
(1) the number of content points (inclusion of differences and similarities)?
(2) the number of reasoned content points (using linguistic devices, i.e., explicatory or defining forms, to support scientific claims)?
(3) the number of content points related to comparative concepts (criteria) (looking at nominalisations, abstract nouns, topic sentences and superordinated concepts)?
(4) the number of lexico-grammatical comparative forms (comparitors)?
These questions align with the framework's core points for comparing (points 2 to 4). The study thus investigates how students co-develop the cognitive, content-oriented and linguistic skills needed to express their emergent knowledge more effectively.
METHOD
Context and Participants
An exploratory study was conducted with 37 lower-secondary (7grade) CLIL biology students (aged 12-13 years), their CLIL biology teacher and their EFL teacher in a charter-like, middle-class school in the north of Spain (Navarra) for three months (January-March 2019). It was decided to work with lower-secondary students, who had just moved from primary to secondary education as this is a decisive stage in students' formal education. The participating school offered biology in three seventh-grade classes with a CLIL modality of 50%, teaching half of its students (approximately 15 per class) in English, who were all included in the study, and half in their L1 (Spanish). These small groups allowed a more personalized teaching approach.
It is a small sample, and cannot be claimed to be representative, but it does offer initial insights into how a group of secondary students learned their science content through a CDF approach. The participants' mother tongue was Spanish (L1), and English was their foreign language (L2), except for three bilingual students (2 Spanish/Chinese and 1 Spanish/Polish). The CLIL biology and EFL teacher ranked the students' English level at a low B1 according to the CEFR.
Design and Method
A class unit was designed in collaboration with the CLIL biology and EFL teacher on the curricular topic of biodiversity, in which the CDF of comparing was integrated into the ordinary science lesson plan, that is, into the class tasks, oral interactions, and materials, providing the students with some CDF-rich lessons. The idea was to integrate the CDF of 'comparing' in a contextualized way so as not to interrupt the regular course of the class or make it look like a subject-ex-
ternal element (see Breeze & Gerns, 2019; Nashaat-Sobhy, 2020; Bauer-Marschallinger, 2022).
We used a pre-and post-test design to measure the possible effects of instruction. Of the three available groups taught by the same CLIL biology and EFL teacher, two were chosen as study groups (S1, S2), and one as control group (CG). The study groups received explicit CDF-based input and output opportunities (see Appendices A-H), while the control group received some lessons only focusing on content aspects without explicit indications or support dedicated to the CDF of 'comparing'; however, as the teacher included comparative expressions in their classroom talk, and as the written texts (see Appendix E) had some comparative lexico-gram-matical and structural features, their classes could be regarded as representing an implicit CDF approach.
Intervention
The intervention included several classes during the first two months of the project (January/February 2019), where the teacher progressively explained the five points of the comparing framework. The first step involved raising students' awareness of the role that comparing plays in science (biology) education and jointly defining what makes a good scientific comparison (see Appendix A). Second, work was done on the first three points of the framework (items being compared, comparative concepts (criteria) and types of comparison) practising it with different scientific topics (see Appendix B). In a third step, the language of and for comparing (comparitors) was introduced, encouraging students to formulate some comparative, content-based statements; here language support was aligned with students' scientific knowledge (see Appendices C and D). Last, some structural devices were introduced, such as parallel and paragraph organisation, and the previous points were revised by deconstructing jointly a written comparison (consult text and tasks in Appendices E-G). As in other studies, a step-wise, joint, in-class learning approach was chosen (see Lose, 2007; Bauer-Marschallinger, 2022; Nashaat-Sobhy, 2020), based on the learning cycle idea from the genre tradition (Rose & Martin, 2012), Cummins's four learning quadrants (2008) from the ESL tradition and Vygotsky's (1934) 'collaborative learning' approach, which recommends moving from more supportive to more autonomous learning. Thus, students'
receptive skills were first activated, followed by their decon-structive and productive ones. The idea was to draw students' attention to the different dimensions that converge when making a comparison, namely the conceptual, cognitive, and linguistic aspects.
Data Collection
To elicit comparisons, we asked the students to hand in three written comparisons: one before and two after the instruction classes, allowing us to trace better students' learning progress and detect any improvement over time. These served as pre- and post-tests (see Table 3). For the pre-test, the students had to compare between herbivores and carnivores, without receiving any kind of guidance or support. Then, during the instruction, the students were asked to rewrite their first written comparison in order to put what they had learned so far during these lessons into practice (see Appendix H). In the post-instruction phase, the students were asked to hand in two more comparisons to examine their ability to process and present new biological content in a comparative way. This time, they had to compare vertebrates with invertebrates and mammals with reptiles. All three tasks followed a similar design based on the previous analysis of 'comparing' and were reviewed by the participating teachers regarding clarity, content authenticity, curricu-lar demands, and level of difficulty.
Data Analysis
The results were measured in quantitative terms by first counting the number of forms used by the individual students in their writings, and then by calculating average for the groups (S1, S2, CG) in each of the three task collection phases (pre-and post-tests). A paired t-test was run with a p-value at 0.05 (Rasinger, 2008), which indicates whether the students' progress was statistically significant. As the sample is small, caution is recommended. The idea is not to present large-scale representative data but to offer first insights into students' CDF content performance and draw some general conclusions on the effects of explicit CDF-teaching.
RESULTS
Table 3
Study Design
CDF-based tasks
Purpose
Schedule
1. compare herbivores and carnivores. (rewrite first comparison)
2. compare vertebrates and invertebrates.
3. compare reptiles and mammals.
Pre-test Instruction Post-test (1) Post-test (2)
January 2019 January/February 2019 Beginning of March 2019 End of March 2019
The quantitative results are presented in the following, and a qualitative pre- and post-test example is provided to illustrate students' content performance better.
Inclusion of Content Points
Starting with the results concerning the inclusion of content points (differences and similarities), the students could present up to 7 points per task. The researcher counted the number of task-relevant points presented by each student, identifying 564 points of differences and 155 of similarities, and calculated the average number per group and phase. The points of difference and similarity were analysed separately.
On the one hand, when considering differences, as shown in Figure 1 and Table 4, all three groups learned to include more points from their pre- to their respective post-tests. In
Figure 1
Mean Number of Content Points (Differences) Used per Student
the pre-tests the students included an average of 4 points, while in their respective post-tests they included an average of 6 points. Based on the paired t-test results (see Table
4), the gains were highly significant for all groups (p< 0.05) except the control group, who improved significantly only when post-test 2 was considered.
On the other hand, regarding students' pre- and post-test inclusion of similarity content points (see Figure 2 and Table
5), at the beginning, the students hardly presented any similarities, whereas after the instructional phase most study-group students (S1, S2) included them, especially in their second post-tests, where they doubled their use (including 2 to 3 forms). This can be considered a significant, delayed improvement (p<0.05) (see Table 5). In contrast, the control group's use of similarities declined.
Content points (inclusion of differences)
pre-test post-test 1 post-test 2
□ Study-group 1 ■ Study-group 2 ■ Control group
Table 4
Paired T-Test: Mean Number of Content Points (Differences) Used between Pre-Test and Post-Tests
Study Variable Groups Pre-Test Post-Test 1 Post-Test 2 P-Value
Mean (Sd) Mean (Sd) Mean (Sd)
Mean number of content points (differences) S1 3.69 (1.58) 5.69 (1.35) < 0.0007 **
S2 3.83 (2.21) 6.25 (2.22) 0.0184 *
CG 3.89 (1.27) 4.89 (2.67) 0.4070
S1 3.69 (1.58) 5.75 (1.61) 0.0010**
S2 3.83 (2.21) 5.92 (2.15) 0.0319*
CG 3.89 (1.27) 5.56 (1.24) 0.0167*
Note. p<*=significance at 0.05; p<**=significant at 0.001
Figure 2
Mean Number of Content Points (Similarities) Used per Student
The following pre and post-test extract (Figure 3, examples 1 and 2) from one study-group student offers a qualitative insight into how most study-group students learned to present the content points in a more integrated, complete, and subject-purposeful way.
As can be seen, in the pre-test (example 1) only one similarity point (both have a skull) is included, which offers little relevant information and is presented separately from the other content points. In the post-test (example 2), the point of similarity offers more task-relevant information and is more integrated within the text. The importance of identifying similarities is recognised as a way of establishing class-memberships, seen by the assertion that herbivores and carnivores share the feature of having a backbone to classify them as vertebrates.
Furthermore, most pre-tests presented the differences in a vague and incomplete way, that is, specifying information about one animal group, while omitting this information for
Post-Test 2
- P-Value
Mean (Sd)
0.8172 0.2562
0.0509
2.94 (1.44) 0.0024 **
2.08 (1.38) 0.0413*
0.44 (0.73) 0.3466
the other. Here, the student explains canines in carnivores but fails to provide the corresponding information about herbivores, indicating, for example, that they have a horny pad instead (herbivores don't have canines and the carnivores have great canines). Something similar can be seen later in example 5 (The carnivores have canins and the herbivores not. The herbivores had large incissors to shop grass and canines no). However, in the post-test most students, like the present one, learned to specify the content points more and analyse the information equally in both groups.
Elaboration of Content Points
Finally, we examined whether the students simply listed or elaborated the content points by providing additional information. The number of content points that were further reasoned, through the use of different linguistic resources (such as synonyms, explanatory/exemplifying forms, conditionals, action verbs, consecutive, relative clauses, etc.), was
Table 5
Paired T-Test: Mean Number Content Points (Similarities) Used between Pre-Test and Post-Tests
Pre-Test
Post-Test 1
Study Variable
Groups
Mean (Sd) Mean (Sd)
S1 1.38 (0.89) 1.31 (1.20)
S2 0.92 (1.31) 1.42 (1.62)
CG 0.78 (0.67) 0.22 (0.44)
S1 1.38 (0.89)
S2 0.92 (1.31)
CG 0.78 (0.67)
Mean number of content points (similarities)
Figure 3
Pre- and Post-Test Example: Presenting Content Points (Differences and Similarities)
(1) PRE-TEST
(2) POST-TEST
PLfftrt*vh
Tk^ Wurbivori*/ toj- CLnd/ fke- u^rnivcnrty
tat wmJ- (difference) aXyo-i^v herbivore-y ticnCj-
cav^Ln&y ajnAy cjxrinZv&re^- h^w^ CM^Sj^vjy. (difference). He-rlpivorcy Pulv^ Ew&tuC ivLCA^wy Guub cajriMsscrrty fwiwHrfs
intMCri*. (difference).
Th^y (jiffiv friO.-ve- OS ¡iuilL (similarity). (8AM)
N&vr, I »» gcn^uj Hr ¡¿U) cUffortMyiy tu^iL
boiK- \u>-vt- a- btwJctwivei i*er fKetj tu-e- \rtrhbro4i-y (similarity). Now, difftrtnciy, rejp^Xty iU-e-
i^oitf^-W^iwUiPl- (e^^o-fkermic)., &w iKe umfa-OAri)
o-<rv wtv^'-btiwiei^ (...)
(difference) (8AM)
Figure 4
Mean Percentages of Elaborated/Reasoned Content Points per Student
Elaborated content points (%)
100 80 60
45 46
34 37 35 35
28 ■ 27 ■ ■
1 20 1
pre-test post-test 1 post-test 2
□ Study-group 1 ■ Study-group 2 ■ Control group
Table 6
Paired T-Test: Mean Value of Elaborated Content Points Between Pre-Test and Post-Tests
Study variable Groups Pre-Test Post-Test 1 Post-Test 2 p-value
Mean (SD) Mean (SD) Mean (SD)
Mean value of content points related to S1 1.38 (1.54) 2.56 (2.53) 0.1308
comparative concepts
S2 1.58 (1.56) 2.67 (2.61) 0.2231
CG 1.22 (1.30) 1.78 (1.56) 0.3251
S1 1.38 (1.54) 3.50 (2.68) 0.0199*
S2 1.58 (1.56) 4.17 (2.17) 0.0040**
CG 1.22 (1.30) 1.69 (0.56) 0.8771
identified and as before, means were calculated in relation to the total number of content points.
Figure 4 and Table 6 show the results of this analysis. When contrasting the pre-test results of all three groups with their respective two post-tests, we see that all three groups improved slightly in developing their content points more. Initially, in the pre-test, the students developed an average of 26-34% of their content points. Then, in the first post-test, these data improved slightly for all, and finally, in the second post-test, the study group learners continued improving significantly on their initial results (see Table 6), whereas the control group decreased the amount of support provided for their content points.
In qualitative terms, the pre-and post-test extracts in examples 3 and 4 (Figure 5) show how especially the study group students learned to elaborate their content points more after the instruction phase.
As the pre-test (example 3) shows, most students presented content points without developing them further or connecting them with theoretical issues. By contrast, in the post-test (example 4), most study-group students improved considerably, clarifying their content points by presenting definitory and explicatory forms. Here, the student explains two biological concepts (cold-blooded and warm-blooded animals) by defining them, using several linguistic resources, such as explanatory and exemplifying forms ('this means'; 'for example'), conditional-if, a consecutive form ('as a result of') and an action verb ('to keep'). The underlying relationship is shown between the different contents (the blood groups, heat regulation mechanism, the environment and eating habits) and there is an attempt to derive a general scientific principle that reptiles are found in warmer climates due to their thermoregulation. This reveals a more complex and interrelated understanding of the contents and of typical scientific ways of proceeding (Schalk et al., 2016).
Figure 5
Pre- and Post-Test Example: Reasoning/Elaborating Students' Comparative Claims
(3) PRE-TEST
(4) POST-TEST
(...) HtrkLv&reA' I-kr- e/it o^ io-h biM-ca^-iK-LvweA'
Wojs WOMM ¿WmWu lïiM' CM'VVLV&Y&S' L^
Uam^ZUS. Herbiwre-^ iWevft^ue ûv bUi a^ut vwy Urvu\
(2CF)
(...) TK^ m^Uvfrv cLLffzveMM* Hato- [4/
iWo^h ve^HJ^/ OJTVs cxAu>U- biorrticd< om^Cl m¿uwm^U^ o^vt' woJrvw- bUrc^UA.. Tk-ti' WVV-XUAJS Huxjt btrtLy
HATV^P^V-OMM'^ crw fii^w <&Mrw<n"un^twh If- L>t Lis
co{Ast iWtw bo^titi/ avz< t^AA. -hrr. Hcrwtsrw, evuuvuvwxl^/ ccwv regt^IWe* fK^w WmA' by -tWewi&W&S'. For ejcawvpLb, iWe-y ge4- coMm- iWt-inuj\^ pwtW: Ay riM/M- of {Wii^ cLofftj'WUsZ'j rßf^HXei/ R^rmiUXy Uat^ uv www /^¿¿LC-e^ btoh wcwvwvHiXy c^uv Liv^ i^v Muj Re^ftX^ n&eds
Le-j^ ftmtL üy ^v^vgy fK^uv mtz^vwuaX^' io-
bottier wc-vv^. (...) (2CF)
Figure 6
Mean Percentages of Content Points Related to Comparative Concepts per Student
Content points related to comparative concepts (0/,
100
80 60 40
20
13
I]
11 11
66 65
pre-test □ Study-group 1
post-test 1 I Study-group 2
post-test 2 ' Control group
Table 7
Paired T-Test: Mean Value of Content Points Related to Comparative Concepts Between Pre-Test and Post-Tests
Study Variable Groups Pre-Test Post-Test 1 Post-Test 2 P-Value
Mean (Sd) Mean (Sd) Mean (Sd)
Mean value of elaborated content points S1 0.50 (0.82) 2.75 (2.44) 0.0051 *
S2 0.42 (0.90) 3.67 (2.84) 0.0059 *
CG 0.44 (0.88) 0.67 (0.87) 0.4468
S1 0.50 (0.82) 3.81 (2.76) 0.0001***
S2 0.42 (0.90) 3.83 (2.72) 0.0009**
CG 0.44 (0.88) 1.89 (1.76) 0.0563
Figure 7
Pre- and Post-Test Example: Using Comparative Concepts (Criteria)
(5) PRE-TEST
(6) POST-TEST
TWe/ Wtrbivorv OMjwaly taA- pUunAy ouMji ib-o to^rni^rorty ta±_ w&aj-, fi^v o~ncL Mgv for i-tuun^Ji/. TKe- caJrniMorty Wo^rv ww» iuvpt ff Wtrbivorty not. TKe- hzrbivorcy luul Uwai-LnciM&rv-to- cWo-f g\ra-yy cwnWorty no:. (11AF) (implicit criteria)
Firyt- i'm- going -fo- tow^wri/ fKe- itgy (explicit criterion) ihaj-ih&y (reptiles and mammals) Wcuvty. TWty uurw be-Itgltvs or ■Mrcbpoda^. LMtw» fKa^-fkey WoAr&nl-t aoi- Lm^ gyub jttrcipodcUs miany jWcuhtf^w itgy.
KtfrtiA&y be/ Itglwor Mra^odaL. Howtvtr wuwvunmXy ewe, ^rofwUU/, iW&y 4 itgy.
S&ooncL /'pw going ftr Wow iyiWtAjr Wabifat
(explicit criterion). BofW uw^ iivty i>v a* HrrtUriaL lund fv e^Rwt^fcv 'At,№. In/ vvuM-ints ew^^tw' onLy c^wv Livt/re^HXty. (11AF)
Content Points related to Comparative Concepts (Criteria)
Third, students' writings were examined on their ability to form comparative concepts (criteria) and base their search for similarities and differences on them. To measure this abstract ability, the number of content points that were explicitly related to a comparative concept using nominalisations, abstract nouns or main idea topic sentences was counted per student and phase, and averages were calculated in relation to the total number of content points (previously measured).
As Figure 6 shows, all post-test results from the three groups indicate a significant improvement (especially the second post-test). In the pre-test, the students only subordinated and introduced a few content points (about 12%), while in the two post-tests, this changed significantly, as the two study groups subordinated 35-54% more forms, and the control group learned to subordinate between 5 to 27% more forms. According to the paired t-test results (see Table 7), this improvement was highly significant for the study-group students (S1, S2) in their two respective post-
tests (p<0.05), while for the control group, the improvement failed to reach significance.
From a qualitative perspective, the following pre- and posttest extracts from a study group student (Figure 7, examples 5 and 6), show the improvement process many study-group students went through, as they learned to relate diffuse content points under some corresponding superordinated concepts (criteria).
In the pre-test (example 5) two animal groups (herbivores, carnivores) are compared on their diet and teeth, but at no point are these two categories explicitly referred to, whereas in the post-test (example 6) the student explicitly names the comparative concepts in a topic sentence ('First I'm going to compare the legs'; 'Second I'm going to explain how is their habitat') and structures his/her comparative analysis on it.
These two examples show that explicit mention of the comparative concept (criterion) in a topic sentence is not an indispensable element when making a comparison, as both the pre-test (example 5) and post-test (example 6) are two valid comparative statements. However, it does show that
Figure 8
Mean Number of Lexico-Grammatical Comparative Forms Used per Student_
L exico - grammati c a 1 comparative forms
12 10 8 6 4 2 0
11,4,
11
5,2
Jit
10,2
pre-test post-test 1 post-test 2
□ Study-group 1 ■ Study-group 2 ■ Control group
Table 8
Paired T-Test: Mean Number of Lexico-Grammatical Comparative Forms Used between Pre-Test and Post-Tests
Study Variable Groups Pre-Test Post-Test 1 Post-Test 2 P-Value
Mean (Sd) Mean (Sd) Mean (Sd)
Mean number of lexico-grammatical S1 4.94 (3.21) 11.38 (5.82) 0.0006 **
comparative forms
S2 4.92 (3.68) 11 (7.38) 0.0145*
CG 3.56 (2.01) 4.44 (2.79) 0.2495
S1 4.94 (3.21) 10.81 (5.56) 0.0005***
S2 4.92 (3.68) 11 (8.05) 0.00981*
CG 3.56 (2.01) 2.78 (2.17) 0.4097
the comparison must have a common comparative point, because otherwise, it would not be a comparison but a description.
Use of Comparative Language (Comparitor)
In the second step, students' writings were analysed in terms of lexico-grammatical comparative forms, the 'com-paritors' (Darian, 2003). There are three main types: contrast forms (such as 'is different', 'unlike'), comparison forms (like 'similar', 'equal to'), and grammatical comparative forms (such as 'less/more than'). In the analysis, all linguistic attempts that somehow indicated a comparative relation were counted, including forms with linguistic errors, as long as their communicative intention was understandable. Average numbers per student, group and phase were calculated. A total of 848 linguistic comparative forms were identified in the whole sample, among which 458 were contrasts, 216 were comparisons, and 174 were grammatical comparative constructions. These numbers can be explained by the fact
that most of the content points presented were differences and there were few similarities.
As Figure 8 shows, the study group students learned to include considerably more lexico-grammatical forms after the instruction classes, progressing from using initially an average of 4 to 5 forms per student to including around 11 in their post-tests, which is a significant improvement (p<0.05) (see Table 8). On the other hand, the control group students also experienced a slight increase between their first and second writing (passing from 3 to 4 forms), but it quickly declined in their second post-test, where they returned to their initial starting point. It should be noted that from the beginning all groups already used some kind of linguistic comparative forms, which may suggest a certain prior familiarization with these forms acquired in their EFL classes.
The following pre- and post-tests (Figure 9, examples 7 and 8), illustrate how most study group students moved from an initial low and basic use of lexico-grammatical expressions (such as 'common'; 'both', 'different', 'but') towards a more
Figure 9
Pre- and Post-Test Example: Using Lexico-Grammatical Comparative Forms (Comparitor)
(7) PRE-TEST
(8) POST-TEST
COV/WMTW.
• (uotiiM^LLy o^d- ke^bwort< (u^jwxLyj kAve. a- iktAL -HwA- proHcA--tb-zir braivu
• haAj-t, a- je-wib&>h iMtdyfa- taA-.
P
• CM-nivOYti- WCArf. tvulii&ry, CdJnjjr^ty, fr-twurlary amt m&Lary Huvh io- taA-.
• Hzrbiwrzy Wavt-: oA-fb-t- begi/pvnimj ib-uj k<ive- H&Ab-, ib-t- kavt- cWe-n^-g wcriwy
p^-twioia-ry
(10CM)
IW n- failc ivbouf fke pb-yiica.1 of bofk- ¡jrinAf^y. 8o-fK- Ki^ve- a/ re^fi/U^-' MTvereii' by
w+vCie- ^LiuvLtnAflXj,' t&Ytrtds by Kftir iWaA-
I^ZMfU-iai f&r b-taA- iKew^eivt^ /tn- So+K-
ki*uiy of ajniwwxiy curt. feAro-tfxrdo-L fb-aA' ib-aA-
■fb-e-y WiWt- fo-\M~ L&gy biM' no-t aX-waAjy, yowxes miUtwiiLl^ LUce- cloi-pWLi^y kitve- fUppvry om-CI ¡,ovmy rejs-WAi' LUoa-yn&k-ty WaArts (MrfCMj Itgy, yo- fk&y cure, LtgLtyy.
So- ltdvisu) l^cWyi^d- IWfs pWy^U-aL vurvr ca»v I iaile abouA- lb-civ ru/>rotLi>-cAioiAs, (W n^re- abovA- you- Ic^ow ib-e- »ej«^ ovifxwov-y a^ui- vwAjtxwov-y. \a1M- iw o-m- Wo^-d- vw-c-wwcXy are- ifwi)WW ttwA- fb-e-y b frennr- jb-e-Lr vvm-wwuxt- kori-y a^-d- ib-ty o-rt- nWytd- wiAb-
mitto frcrvw wcAWt-r gio^dy, ¿hx^* L**- ib-e- otb&r b-c^d ri^>HXt-y ojrty vWifXirov-y, fWtij Lo-y tggy uv fb-e- La*u>U
(10CM)_
varied and combined use of advanced forms (such as 'while', 'whereas', 'on the one hand... on the other'), making their comparative claims more explicit.
The students had no great difficulties in picking up and using these forms; they probably knew them from their EFL class and they seemed to understand their functional purpose as they used them in the right place (to state a contrast and/or similarity). However, they had difficulties using them correctly, that is without making lexico-grammatical or orthographic mistakes. For example, they confused comparatives with superlatives, forgot about the suffix 'than' when using comparative grammatical forms or used incorrect word order or prepositions, which suggests inadequate grammatical, lexical and orthographic knowledge or perhaps lack of attention due to cognitive overload.
Moreover, as shown in example 8, the students were able to combine these forms with other linguistic devices, which had not been explicitly taught, which were probably transferred from their L1 and/or EFL classes. These include, consequence forms ('so'), the English possessive 's', explicative and defining forms (such as 'that means that', 'that is', 'like'), quantifiers ('some', percentages), and they developed some awareness to the need to guide readers and adopt a writerly stance, using future tenses, direct pronouns ('I', 'you'), discourse markers ('well', 'so'), temporal features ('now') and referring to a shared background knowledge ('I am sure about you know the meaning of ...'). Their posttests thus show greater literacy development by decentring (take distance) from the text and establishing a dialogue with the reader.
DISCUSSION
This paper has examined the teachability and effect of explicitly teaching CDFs in CLIL science.
As for the teachability of CDFs, the CLIL science and the English teacher had no difficulty in co-designing and delivering science lessons that integrated the CDF of comparing. They quickly understood that language not only refers to English (as L2) but to the language demands specific to science, which goes beyond learning a set of technical terms, but includes using abstract nouns and topic sentences to form concepts, supporting scientific claims through linguistic devices and using discursive structures and specific lexical-grammatical forms to express comparative relationships (Lemke, 1990; Mortimer & Scott, 2003; Evnitskaya & Morton; 2011; Rose & Martin, 2012; Nikula, 2017). A functional rather than formal approach was adopted to the study of language and content, where clarity (communicative intention) was favoured over form. That is why little attention was paid to grammatical and orthographic errors, so as not to overload students with corrections and avoid turning the class into a language class.
The intervention showed that teaching students how to compare from a scientific perspective and in a progressive, explicit and interactive way was extremely useful. A gradual learning path was chosen, which started by first co-defining the importance of comparing for learning natural sciences, then moved on to practise some of its constituent components, to review lexico-grammatical forms for expressing the emerging scientific knowledge better and finally to ad-
dress textual organization in paragraphs and parallel structures. The language and cognitive support were only introduced in so far as they aided subject learning.
As for the effect of explicitly teaching CDFs on students' subject performance, significant results were obtained for the experimental groups, which improved in both content and language learning, when looking at the inclusion of content points, justification of scientific claims, concept formation and use of lexico-grammatical forms.
Based on the results, the study-group students moved from an initial, weak understanding of comparing towards a more proficient one as students learned to work harder on content and language. They included a greater number not only of differences but also of similarities and provided more specific, and task-relevant information. They also explained and linked their initially diffuse content points more, relying on linguistic devices (such as linking words, defining and explicatory forms) and clarifying the relationship between content concepts (i.e., blood type, environment and temperature), showing a clearer subject understanding. Moreover, the students also learned to structure their claims and make them more explicit as they related the contents to the superordinate concepts, introduced them with topic sentences and signalled comparative relationships using appropriate lexico-grammatical forms, indicating a more advanced linguistic command and proficiency in abstract reasoning. By and large, the explicit instruction helped students reach better scientific achievement levels as they developed more scientific concepts, practised scientific thinking (i.e., using similarities to establish class membership) and produced more sophisticated writing about science. The students understood that comparing in science is an inherent part of studying science, requiring certain cognitive and linguistic forms. The process of writing in science class, which is not frequently used, undoubtedly supported the students' learning process, helping them to plan, organise and visualise their scientific understanding in a linear and logical way (Schleppegrell & Colombi, 2005; Oliveira & Weinburgh, 2017).
Most of the pre-test results coincide with the findings made by other studies, which observed students' natural - that is, not-explicitly-taught - CDF performance in different predominantly secondary-level but also tertiary European CLIL contexts, (see Lose, 2007; Coetzee-Lachmann, 2009; Breeze & Dafouz, 2017; Lorenzo, 2017; Dalton-Puffer et al., 2018; Evnitskaya & Dalton-Puffer, 2020; Whittaker & McCabe, 2023). They show that students have difficulties performing CDFs when left on their own, often presenting them in an incomplete way, without including many concepts or explaining them in depth. Such was the case in the pre-test phase, where students failed to present the contents accurately and completely, doing so in a somewhat superficial way, lacking solid argumentation and supporting information. In
addition, the students seem to prefer to focus on concrete, tangible content elements rather than relating them to abstract concepts (i.e., comparative criteria), a phenomenon that Evnitskaya and Dalton-Puffer (2020) have analysed in terms of horizontal and vertical content performance. As Nashaat-Sobhy (2020) explains, vertical performance of this kind is particularly challenging.
Overuse of horizontal content performance may be linked to weak conceptual and procedural knowledge, and an inadequate understanding of how knowledge should be displayed using CDFs (Breeze & Dafouz, 2017; Evnitskaya & Dal-ton-Puffer, 2020). Clear proof of this is that the students at first only superficially included similarity points, to comply with the task instruction, and failed to name the comparative concepts (criteria). It seems that if students are left on their own without explicit support, they have considerably more difficulty presenting the content points in a subject-purposeful way.
The post-tests are consistent with recent interventional studies (Connolly, 2019; Breeze & Gerns, 2019; Nashaat-Sob-hy, 2020; Bauer-Marschallinger, 2022; Hasenberger, ongoing), showing how even general, non-explicitly focused CDF support can be very helpful in equipping students to display their content knowledge better, as was the case with the control group who received implicit CDF-based classes. When contrasting the study and control group's CDF-performance, interestingly, both groups did improve in terms of including differences and building comparative concepts (criteria). However, the control group did not progress in adding similarities, using lexico-grammatical forms or justifying their statements on their own. An implicit CDF-teaching approach seems to help students pick up some CDF-related features, probably the more familiar ones. In this sense, Breeze and Gerns (2019) reported improvements in a group of Spanish secondary CLIL history students who worked on academic writing skills involving the CDFs of 'describing' and 'explaining', leading to a higher number of content points and improved structure.
In more explicit studies, where the CDF of 'defining' was taught in Spanish tertiary education (Nashaat-Shobhy, 2020), the CDF of 'explaining' in German bilingual secondary chemistry classes (Connolly, 2019) and all seven CDFs were taught in Austrian CLIL secondary history and science classrooms (Bauer-Marschallinger, 2022; Hasenberger, ongoing), a reciprocal effect could be observed on students' content and language learning. It seems that teaching with CDFs helps students retain their subject knowledge better (Na-shaat-Sobhy, 2020), increases their use of CDF-related expressions and linking devices (Bauer-Marschallinger, 2022; Hasenberger, ongoing) and fosters their self-confidence (Connolly, 2019). The present post-test results confirm that an explicit teaching approach has a stronger effect on stu-
dents' integrated content and language learning, making them focus more on linguistic and structural features (such as comparitors, linking devices/sequential connectors, explanatory and defining forms, topic sentences, paragraph, and parallel organisation among others), which they would most probably not pay attention to if not prompted.
The results also show how teaching science through a CDF approach may require more time and practice since some points of the framework developed over a longer period (see delayed improvement in presenting similarities in the second post-test).
CONCLUSION
The CDFs construct offers content teachers a practical framework to unlock the importance of language and cognition in the teaching and learning of natural sciences, two dimensions that often receive little attention. This integration can be promoted in a natural way, since the content teacher can point out certain linguistic or cognitive aspects from the scientific perspective without resorting to more complex language models from the EFL class. One of the advantages of the CDFs is that they are basic knowledge-building blocks common to all school subjects (Morton, 2020) and thus inherent to science education.
Furthermore, the study demonstrates the integrated learning potential of CDFs, which can equip students to deal better with content, cognitive and language aspects in a CLIL science class. The CDFs help students advance in their conceptual knowledge and its corresponding language, something students are expected to develop during their years of schooling, which they will not learn in their language class or outside school. These functions, therefore, offer teachers a good starting point to introduce students to subject-specific ways of making meaning and communicating it appropriately, enabling them to participate in scientific discourse in and beyond school. It would thus be desirable that con-
REFERENCES
tent teachers include more of these cognitive functions in their class input and output to help students in their knowledge-building process. For this, a gradual and collaborative learning path is recommended, making the conceptual, cognitive and linguistic demands visible to students, as most require support to notice and use these tools.
Some of the limits of the present study were the small sample size and the lack of variety between the different written tasks, which follow a similar task design based on a comparative analysis. It would have also been interesting to analyse teacher's and students' classroom interaction and have taught the CDF of comparing in another science class (e.g., chemistry).
Further research is therefore needed to operationalize the different CDFs across school subjects and levels, and to provide practical pedagogical tools to foster students' scientific knowledge in different CLIL contexts.
ETHICS STATEMENT
This study formed part of a research project, which was accepted by the Research Ethics Committee (CEI) of the University of Navarra (approval number 2019.164).
ACKNOWLEDGEMENTS
The author is grateful to the teachers and students who participated in the study and also wishes to thank the Fundación Ciudadanía y Valores (Funciva) for their generous support.
DECLARATION OF COMPETITING INTEREST
None declared.
Anderson, L. W., & Krathwohl, D. R. (2001). A taxonomy for learning, teaching, and assessing: A revision of Bloom's taxonomy of educational objectives. Longman.
Badertscher, H., & Bieri, T. (2009). Wissenserwerb im content and language integrated learning [Knowledge acquisition in content and language integrated learning]. Haupt.
Bauer-Marschallinger, S. (2022). CLIL with a capital I [Unpublished Doctoral dissertation]. University of Vienna.
Biggs, J. B., & Tang, C. S. (2011). Teaching for quality learning at university. Open University Press.
Bloom, B. S. (1956). Taxonomy of educational objectives, Handbook I: Cognitive domain. Longman.
Breeze, R., & Dafouz, E. (2017). Constructing complex cognitive discourse functions in higher education. System, 70, 81-91. https://doi.org/10.1016/j.system.2017.09.024
Breeze, R., & Gerns Jimenez-Villarejo, P. (2019). Building literacies in secondary school history: The specific contribution of academic writing support. EuroAmerican Journal of Applied Linguistics and Languages, 6(1), 21-36. https://doi. org/10.21283/2376905X.10.149
Breeze, R., Llamas Saiz, C., Martinez Pasamar, C., & Tabernero Sala, C. (2014). Integration of Theory and Practice in CLIL. Rodopi/ Brill.
Carpi, A., & Egger, A. E. (2011). Comparison in scientific research. In A. Carpi & A. E. Egger (Eds.), Process of science (pp. 129-142). Visionlearning.
Cheong, L. K. (1978). Syntax of scientific English. Singapore University Press.
Clark, S.K. Judd, E., Smith, L.K., & Ahlstrom, E. (2020). Examining the effects of integrated science and literacy instruction to teach second graders to write compare and contrast informational texts. Early Childhood Education Journal, 107(3), 567-579. https://doi.org/10.1007/s10643-020-01106-9
Coetzee-Lachmann, D. (2009). Assessment of subject-specific task performance of bilingual geography learners [Unpublished doctoral dissertation]. University of Osnabrück.
Connolly, T. (2019). Die Förderung vertiefter Lernprozesse durch Sachfachliteralität [The promotion of in-depth learning processes through subject literacy] [Doctoral dissertation]. Johannes Gutenberg University Mainz.
Cummins, J. (2008). BICS and CALP: Empirical and theoretical status of the distinction. In N. H. Hornberger (Eds.), Encyclopedia of Language and Education (pp. 487-499). Springer. https://doi.org/10.1007/978-0-387-30424-3_36
Coyle, D., & Meyer, O. (2021). Beyond CLIL: Pluriliteracies teaching for deeper learning. Cambridge University Press.
Coyle, D., Meyer, O., & Staschen-Dielmann, S. (2023). A deeper learning companion for CLIL: Putting pluriliteracies into practice. Cambridge University Press.
Dalton-Puffer, C. (2004). Academic language functions in content and language integrated classrooms. Vienna English Working Papers, 13 (1), 23-48.
Dalton-Puffer, C. (2007). Academic language functions in a CLIL environment. In D. Marsh & D. Wolff (Ed.), Diverse contexts -converging goals (pp. 201-210). Peter Lang.
Dalton-Puffer, C. (2013). A construct of cognitive discourse functions for conceptualizing content-language integration in CLIL and multilingual education. European Journal of Applied Linguistics, 1(2), 216-253. https://doi.org/10.1515/eujal-2013-0011
Dalton-Puffer, C., & Bauer-Marschallinger, S. (2019). Cognitive discourse functions meet historical competences: Towards an integrated pedagogy in CLIL history education. Journal of Immersion and Content-Based Language Education, 7(1), 30-60. https://doi.org/10.1075/jicb.17017.dal
Dalton-Puffer, C., Bauer-Marschallinger, S., Brückl, K., Hofmann, V., Hopf, J., Kröss, L., & Lechner, L. (2018). Cognitive discourse functions in Austrian CLIL lessons. EuropeanJournal of Applied Linguistics, 6(1), 5-29. https://doi.org/10.1515/eujal-2017-0028
Darian, S. (2003). Understanding the language of science. University of Texas Press.
DeBoer, M., & Leontjev, D. (Eds.). (2020). Assessment and learning in content and language integrated learning (CLIL) classrooms. Springer. https://doi.org/10.1007/978-3-030-54128-6
De la Paz, S., & Mc Cutchen, D. (2010). Learning to write. In R. E. Mayer & P. A. Alexander (Eds.), Handbook of research on learning and instruction (pp. 32-54). Routledge.
Del Pozo, E., & Llinares, A. (2021). Assessing students'learning of history content in Spanish CLIL Programmes. In International Perspectives on CLIL (pp. 43-61). Palgrave Macmillan Cham.
Dixon, R.M. (2005). Comparative constructions in English. Studia Anglica Posnaniensia, 41, 5-27.
Doiz, A., & Lasagabaster, D. (2021). An analysis of the use of cognitive discourse functions in English-medium history teaching at university. English for Specific Purposes, 62, 58-69. https://doi.org/10.1016/j.ESP.2020.12.002
Evnitskaya, N., & Dalton-Puffer, C. (2020). Cognitive discourse functions in CLIL classrooms. International Journal of Bilingual Education and Bilingualism, 1-20. https://doi.org/10.1080/13670050.2020.1804824
Evnitskaya, N., & Morton, T. (2011). Knowledge construction, meaning-making and interaction in CLIL science classroom communities of practice. Language Education, 25, 109-127. https://doi.org/10.1080/09500782.2010.547199
Flannery, M. C. (2010). Biology today: Compare, compare. The American Biology Teacher, 72(6), 378-381. http://dx.doi.org/10.1525/abt.2010.72.6.13
Gerns, P. (2021). Cognitive discourse functions in content and language integrated learning: An empirical study in secondary school science [Unpublished doctorl dissertation]. University of Navarra.
Gerns, P. (2023). Qualitative insights and a first evaluation tool for teaching with cognitive discourse function: "comparing" in the CLIL science classroom. Porta Linguarum An International Journal of Foreign Language Teaching and Learning, (40), 161-179. https://doi.org/10.30827/portalin.vi40.26619
Goldstone, R. L., Day, S., & Son, J.Y. (2010). Comparison. Springer.
Halliday, M.A.K., & Hasan, R. (1976). Cohesion in English. Longman.
Halliday, M.A.K., & Matthiessen, C.M.I.M. (2004). Halliday's introduction to functional grammar. Routledge.
Hammann L.A, & Stevens R. J. (2003). Instructional approaches to improving students' writing of compare-contrast essay. Journal of Literacy Research, 35(2), 731-756. https://doi.org/10.1207/s15548430jlr3502_3
Hasenberger, T. (n.d.). English for the natural sciences: Developing and implementing a curriculum for a new subject at upper-sec-ondaryschools [Unpublished doctoral dissertation]. University of Vienna.
Huddleston, R. (2017). Comparative constructions. In R. Huddleston & G. K. Pullum (Ed.), The Cambridge Grammar of the English Language (pp. 1097-1170). Cambridge University Press.
Lemke, J. L. (1990). Talking science. Ablex Pub. Corp.
Llinares, A. (2015). Integration in CLIL: A proposal to inform research and successful pedagogy. Language Culture and Curriculum, 28(1), 58-73. https://doi.org/10.1080/07908318.2014.1000925
Llinares, A., & Dalton-Puffer, C. (2015). The role of different tasks in CLIL students' use of evaluative language. System, 54, 60-79. https://doi.org/10.1016ij.system.2015.05.001
Llinares, A., Morton, T., & Whittaker, R. (2012). The roles of language in CLIL. Cambridge University Press.
Llinares, A., & Nashaat-Sobhy, N. (2021). Defining science in primary school CLIL contexts. Language Teaching for Young Learners, 3(2), 337-362. https://doi.org/10.1075/LTYL.20010.LLI
Llinares, A., & Nikula, T. (2023). CLIL students' production of cognitive discourse functions: Comparing Finnish and Spanish contexts. Language and Education. https://doi.org/10.1080/09500782.2023.2211049
Lorenzo, F. (2017). Historical literacy in bilingual settings. Linguistics and Education, 37, 32-41. https://doi.org/10.1016/jJin-ged.2016.11.002
Lose, J. (2007). The language of scientific discourse. In D. Caspari, W. Halle, A. Wegner, & W. ZydatiR (Ed.), Bilingualer unterricht machtschule [Bilingual education makes school] (pp. 97-107). Peter Lang.
MacArthur, C. A., & Philippakos, Z. (2010). Instruction in a strategy for compare-contrast writing. Exceptional Children, 76(4), 438-456. https://doi.org/10.1177/001440291007600404
Martinez, P. (2018). The comparative method in biology and the essentialist trap. Frontiers in Ecology and Evolution, 6, 130. https://doi.org/10.3389/fevo.2018.00130
Marzano, R. J. (2001). Designing a new taxonomy of educational objectives. Experts in assessment. Corwin Press.
Mercer, N. (1995). The guided construction of knowledge. Multilingual Matters.
Meyer, O., Coyle, D., Halbach, A., Schuck & K., Ting, T. (2015). A pluriliteracies approach to content and language integrated learning - mapping learner progressions in knowledge construction and meaning-making. Language, Culture and Curriculum, 28(1), 41-57. https://doi.org/10.1080/07908318.2014.1000924.
Meyer, O., Coyle, D., & Schuck, K. (2018). Learnscaping: Creating next-gen learning environments for pluriliteracies growth. In J. Buendgens-Kosten & D. Elsner (Eds.), Multilingual computer assisted language learning (pp. 18-38). Multilingual Matters.
Mortimer, E.F., & Scott. P.H. (2003). Meaning making in secondary science classrooms. Open University Press.
Morton, T. (2020). Cognitive discourse functions. CLIL Journal of Innovation and Research in Plurilingual and Pluricultural, 3(1), 7-17. https://doi.org/10.5565/rev/CLIL.33
Nashaat-Sobhy, N. (2020). Operationalizing 'defining' from a cognitive discourse perspective for learners' use. In S. M. An-waruddin (Eds.), Knowledge mobilization in TESOL (pp. 94-112). Brill Sense. https://doi.org/10.1163/9789004392472_007
Nikula, T., Dafouz, E., Moore, P., & Smit, U. (2016). Conceptualising integration in CLIL and multilingual education. Educational Linguistics (vol. 32). Springer.
Nikula, T. (2017). 'What's the Moment Thingy?'. In J. Langman, & H. Hansen-Thomas (Ed.), Discourse Analytic Perspectives on STEM Education (pp. 11-29). Springer International Publishing. https://doi.org/10.1007/978-3-319-55116-6_2
Norris, S., & Phillips, L. (2009). Scientific literacy. In D. Olson & N. Torrance (Eds.), The Cambridge handbook of literacy. Cambridge University Press.
Oliveira, A. W., & Weinburgh, M. H. (2017). Science teacher preparation in content-based second language acquisition. Springer.
Polias, J. (2015). Apprenticing students into science. Lexis Education.
Raphael, T. E., & Kirschner, B. M. (1985). The effects of instruction in compare/contrast text structure on sixth-grade students' reading comprehension and writing products. Research Series (vol. 161). Michigan State University.
Rasinger, S. M. (2008). Quantitative research in linguistics: An introduction. Continuum.
Roberts, D.A., & Bybee, R. W. (2014). Scientific literacy, science literacy, and science education. In N. Lederman & S. Abell (Eds.), Handbook of research on science education. Routledge.
Rose, D., & Martin, J. R. (2012). Learning to write, reading to learn: Genre, knowledge, and pedagogy in the Sydney school. Equinox Publishing.
Salvador-Garcia, C., & Chiva-Bartoll, O. (2022) The nature of integration in physical education through CLIL. Journal of Language, Identity & Education. https://doi.org/10.1080/15348458.2022.2085107
Schalk, L., Saalbach, H., & Stern, E. (2016). Approaches to foster transfer of formal principles. PloS One, 11(2), e0148787. https://doi.org/10.1371/journal.pone.0148787
Schleppegrell, M.J., & Colombi, M.C. (2005). Developing advanced literacy in first and second languages. Routledge.
Silver, H. F. (2007). Compare & contrast: Teaching comparative thinking to strengthen student learning. Association for Supervision & Curriculum Development.
Smith, S. (2019). Academic writing genres. Evident Press.
Vanderbeke, M., & Wilden, E. (2017). Sachfachliche Diskursfähigkeit durch fremdsprachliche affordances in bilingualen Schülerlaborprojekten [Subject-related discourse skills through foreign language affordances in bilingual school laboratory projects]. Zeitschrift für Fremdsprachenforschung, 2(1), 3-27.
Veel, R. (1997). Learning how to mean - scientifically speaking. In F. Christie & J. R. Martin (Eds.), Genre and Institutions (pp. 161-195). Continuum.
Villabona, N., & Cenoz, J. (2021). The integration of content and language in CLIL. Language, Culture and Curriculum, 35, 36 - 50. https://doi.org/10.1080/07908318.2021.1910703
Vollmer, J. H. (2010). Items for a description of linguistic competence in the language of schooling. Language Policy Division.
Whittaker, R., & A. McCabe. (2023). Expressing evaluation across disciplines in primary and secondary CLIL writing. International Journal of Bilingual Education and Bilingualism, 26(3), 345-362. https://doi.org/10.1080/13670050.2020.1798869.
Widdowson, H. G. (1979). Reading and thinking in English. Oxford University Press.
Zwiers, J. (2014). Building academic language: Meeting common core standards across disciplines. Jossey Bass Teacher.
APPENDIX A
'Comparing' in science
Why do we compare in biology? What is important forwritinga good comparison in science?
APPENDIX B
Components of the CDF 'comparing'
APPENDIX C.
The language of and for comparing (lexico-grammatical forms; comparitor)
APPENDIX D
Individual written tasks
Please form sentences that indicate how herbivores and carnivores are different. To do this, use the contrast forms of the box. Do it as the example indicates.
Example:
1. Herbivores are plant eaters, while carnivores are meat-eaters.
2. Herbivores have a tooth set that is very different from carnivores in...
3.
4.
5.
6.
Expressions for contrast
is different from/ unlike An herbivore can be distinguished from a carnivore in... is not as (simple) as
An herbivore is different from a carnivore in.... Or: Different from/Unlike (herbivores), carnivores...
In contrast to/In comparison to/ compared/ opposed to (herbivores), carnivores...
herbivores and carnivores are different/dissimilar in ....
j on the one hand
but/
herbivores..., whereas/however/ carnivores... on the other hand
One /another difference between] herbivores and carnivores is/lies on the ... Linking words:
• but, whereas, while, although, on the contrary, instead.
APPENDIX E
Reading and writing tasks
1. Read the following text with your neighbour. Identify comparative SIGNAL WORDS. Circle the ones indicating a similarity in green and the ones indicating a difference in red. Fill these expressions in the table below.
Signal words
similarity difference
2. In the text, go paragraph by paragraph and identify the main idea on which the two animals (herbivores and carnivores) are compared. In the first paragraph, for example, carnivores and herbivores are compared on their diet. Now try to figure out what these main comparative points are for the other paragraphs.
3. Find out what the text says about the concrete differences and similarity. What do the two groups have or do not have in common? Include this information in a Venn diagram or a table. Good luck! ©
Type of
food/ Diet
Differences and similarities between herbivores and carnivores
In the animal kingdom, two major kinds of animals can be distinguished. These include herbivores and carnivores that will be compared.
1.Herbivores, as well as carnivores, are similar in being heterotrophs. This means both animal groups depend on others for food. They require another living thing to live. But, having a closer look at the concrete type of food they eat, they differ. Herbivores refer to those animals that consume only plants. This includes leaves1, grains2 and seeds3. In contrast, carnivores are those animals that eat meat to survive4.
2.Determining whether a particular animal is a herbivore or a carnivore is not just limited to merely observing what particular type of food they eat, but by simply looking at the skeletal remains5: at their teeth. Both groups possess an identifiable tooth set that allows them to eat. On the one hand, herbivores have broad, flat canines6, sometimes they are even absent, and with large, dull incisors8. This helps them to properly grind9 the fibres found in plants and crush seeds and grains. On the other side, carnivores have teeth proper to a hunter with large, pointed canines and short, sharp incisors. This allows them to tear10 through skin and muscle and chew11 the meat from their prey12.
3. Another comparable feature is the nails. Herbivores have flattened nails13 or blunt hooves14. Carnivores have sharp claws15instead. Again, the difference between the nails has to do with the type of food they eat. Carnivores have to hunt and kill their prey to eat their flesh, and these claws help them leave their prey helpless.
4. In line with this, there is also a difference in the size17 of the mouth opening with the size of the skull18. As herbivores only eat plants, their mouth opening is relatively small compared with their head size. In contrast, carnivores have larger mouth openings. This is because carnivores do not just use their mouths to eat but utilize their sharp teeth to attack and kill their prey.
5. A last point to be compared is the digestive system of both groups. As digestion starts in the mouth, by breaking down the food into smaller pieces, in both cases, their teeth are indispensable for this process. Nevertheless, digestion is not as complex in carnivores as in herbivores. Carnivores have a very simple, single-chambered stomach because the meat is relatively easily digested. Hence19, their small intestine is quite short too. On the contrary, herbivores that consume high amounts of plants, difficult to digest must do a lot of chewing. That is why they have a complex, multi-chambered stomach and a long small intestine.
Concluding, herbivores only utilize their flat-based, dull teeth for consuming food (plant leaves, seeds and grains) which is why they only have small mouth openings. While carnivores use their sharp incisors and canines as a weapon20, making it necessary for their mouth openings to be relatively larger.
APPENDIX F
Examples of students' responses
Characteristic:
Diet/Type of food
Teeth Houth,
Group 1: Herbivores
Herbivores: eat plants (learn, grains, seeds, ec{l Carnivores: eatmeat, HwtwaesjSf&LiteAôi rotfe HertùicxeE,., ^^
UXljjÛfS
ma-tH-CÇÛAW^
Herbwctfes o <£#np0ey
motStc - LWimbeflïd iptymciiJh.
&m ¡xieir
Group 2: Carnivores
Gôrrùjafâbï ishcvfp
petnbad canme^. sharp -noÄüfs
Comjuores
OpouYig iarnCvcnsbS bCmfôe
sèuiTtxcK
ërtfë. yra® ùitei ¿ne
Common (=)
fejth. cxre heteititepK
ta both du^jxori stofts en the
ïr.Djth
APPENDIX G
General structure of a comparison and useful starting sentences
Useful phrases:
1. Introduction... What is going to be " compared?
This text will compare and contrast...
In the . there are two major kinds of animals:... and... thai will be compared. What makes a (carnivore) different from an (herbivore)? Does only the (food diet) distinguish them? It will follow an analysis of both animal groups.
2. Middle part... Similarities & Diferences of both groups
3.Conclusion/End A summarising statement.
One first difference is that.../ First of ail/To begin with... Another difference is.../There is also a difference in... Next/ Besides/ In addition/ Furthermore/ Second ... (For instance, for example, such as/ In other words, namely,)
To conclude/ Finally /In the end/ Concluding... To sum up/summerasing/In short/Altogether ...
APPENDIX H
Useful framework to redo your first comparison
Remember!! Steps for writing a scientific comparison:
1. Identify the groups, that will be compared.
2. Make a brainstorming on characteristics both groups have in common
(similarities) or not (differences). (Do not list irrelevant information!) (Note
them in a Venn diagram or tablet).
3. Order the points logically.
4. Time for writing.
Give your text a structure!
(1. Introduction^. Middle part (Point-by-point Method), 3. Conclusion).
Start with an introduction.
5. Link isolated ideas and use signal words to stress similarities and
differences.
6. Do not forget a summarising conclusion! ©
