Could competence-based chemistry teaching in secondary school harm students’ performance in upper traditional exams?

Abstract

In this research, we have explored the possible differences in students’ performance in grade nine chemistry exams that could have been influenced by a different instructional approach in grade eight. We compared two groups of students: one group had received a transmissive propaedeutic approach in grade eight (especially related to the memorising the Periodic Table and mastering the rules of chemical nomenclature), and another group had received competence-based instruction in grade eight (more emphasis on modelling and inquiry activities). Both groups were homogeneously mixed at the end of grade eight, and they both received a transmissive approach in grade nine. Data were gathered through four exams over grade nine in both groups, and final course grades were also retrieved and compared. We found significant differences in assessment scores between both groups at the beginning of grade nine, but those differences were not found in subsequent assessments and in students’ final grades. Therefore, our results would suggest a scarce effect of the instruction of formal and propaedeutic content in grade eight, as all students could reach the same level regardless of the instruction received in the previous year.

---

Introduction

Competence-based education is perhaps one of the greatest revolutions in education experienced in recent decades. Within the global competence framework, the Programme for International Student Assessment, or PISA programme (OECD, 2019, p. 99) defined three specific competences for scientific literacy. These can be summarised as the capacity to (a) explain phenomena scientifically, (b) evaluate and design scientific enquiry; and (c) interpret data and evidence scientifically. The competence framework has influenced core curricula in many parts of the world. In the case of Catalonia (Spain), it has spurred an in-depth change in national curricula (Sanmartí, 2020). This pedagogical shift has prompted an intense debate within the local educational community, with a wide range of visions and perceptions: from those teachers who enthusiastically adopted the competence framework to those who do not feel confident in their ability to vary instructional strategies (Schleicher, 2020), or those who perceive it as a backsliding in education (Zhang, 2016).

Within the spectrum of reluctant teachers, some voices express their fear (well-grounded or not) that the shift towards competence-based instruction may limit the advancement of those students who want to continue on to higher education (Biesta et al., 2015). It is common that these reluctant voices believe that competence-based instruction in the early grades of secondary education might negatively affect instructional programme coherence (Newmann et al., 2001) and, therefore, harm students’ performance in upper high school traditional exams. Particularly, it is common for secondary school science teachers in Catalonia to debate whether they should focus on developing the scientific competences presented above or, by contrast, on preparing students for “harder” and more formal content which will be essential in their academic futures (that is, propaedeutic content), such as learning chemical nomenclature, memorising the periodic table, or solving mathematical-based chemistry problems (Kazempour, 2009). In fact, it is well known that teachers’ beliefs strongly effect how the global educational frameworks and innovations are translated into specific teaching practices (Cheung and Ng, 2000; Zuber and Altrichter, 2018). These beliefs are often related to the difficulties facing secondary school science teachers as they try to apply a competence-based vision to their programmes such as, for example, not considering themselves capable of mastering difficult situations and a lack of training (Zuber and Altrichter, 2018, p. 196). For this reason, even though the competence framework in science education emerged decades ago, the actual translation of this framework into secondary school science classroom practices is far from being a reality, both in Catalonia and abroad. We summarise the main discrepancies found in the design of chemistry teaching regarding three main aspects: (a) the role of the student, (b) the role of the content and (c) the role of assessment.

Regarding the teacher and the role of students

Students’ different roles in the classroom are one outcome of teachers’ main pedagogical beliefs, as described by Cheung and Ng (2000), Luft and Roehrig (2007) and Serin (2018). On the one hand, methodological approaches that are focused on conveying the teacher's knowledge to their students, also called transmissive or teacher-centred methodologies, can be found in which an eminently passive role for students is implied (Serin, 2018). The beliefs that sustain these methodologies are that the student has little or no knowledge of the subject of study, and that if it is explained clearly and the learner pays attention and has sufficient capacity, they will be able to store the necessary content in their memory (Cheung and Ng, 2000). On the other hand, other approaches can be found that consider learning a personal and social activity, as described by Luft and Roehrig (2007) and Serin (2018), such as learning sciences in context, problem-based learning, project-based learning, inquiry-based or modelling-based learning, and others. These student-centred methodologies, which require students to play an active, leading role in order to achieve the learning objectives, draw on the belief (and the consequent pedagogical knowledge) that what students think, their preconceptions and their prior ideas, are the starting points of the instruction needed (Serin, 2018).

Regarding the teacher and the role of the content

From our perspective, another relevant issue derived from differences in teachers’ pedagogical beliefs described in previous studies, such as in Cheung and Ng (2000) and Luft and Roehrig (2007), revolves around what content is chosen for students to learn and how they are arranged throughout instruction, as in many contexts national curricula only provide general guidelines about what to learn and how. For some teachers, it is important to teach many contents which are small parts of the theoretical models of science, without showing the relations among them. In these situations, they usually start with a high degree of abstraction and complexity and seldom connect them to the phenomena occurring around the students every day. Furthermore, these contents are repeated in different academic years with few changes, as many students have neither understood nor internalised them (Cokelez, 2012; Johnson, 2005). Oftentimes, the content where the students might have the most difficulties is introduced even if it is not included in the official curricula for this class (propaedeutic content), and it is practised every year. In contrast, other approaches advocate choosing the knowledge to be constructed around a few, big and core ideas, such as “all matter in the Universe is made of very small particles” (Harlen, 2015, p. 15) and increasing the level of complexity of each idea across the different academic years. These approaches need to rely on a detailed study of how students can potentially move toward a more sophisticated understanding of those core ideas over a broad defined period of time (Stevens et al., 2010). In some cases, the development of students’ scientific ideas has been characterised through learning progressions, which are model pathways of learning over periods of time (Duncan and Rivet, 2013; Duschl et al., 2011). However, as there are not always detailed learning progressions available for all core ideas, we argue that this lack of educational guidelines can hamper the development of instructional programmes for teachers who want to follow this educational approach.

This debate between teaching a long list of scientific content and constructing few but core scientific ideas is of particular interest for chemistry education and the development of the core ideas about the structure of matter. It is fairly well established that novice learners assume that matter is continuous, with no underlying structure (Talanquer, 2009; Tsitsipis et al., 2012). In middle school or lower secondary education, students should start building a basic particle model for matter to explain real-world phenomena (e.g., smells travelling across the room), involving the concept that all matter is made up of particles that are in constant, random motion (Stevens et al., 2010, p. 699). Although at this first stage there is no need for a more complex model than a sphere when explaining phenomena (Stevens et al., 2010), students might face difficulties when developing these ideas, such as thinking that the particulate model of matter means that matter is granular, either thinking of a substance made up of little pieces of the same material or as a continuous entity with embedded particles of some generic kind (Talanquer, 2009). Overcoming this conceptual gap between a phenomenological approach to matter and a particulate conception of matter represents one of the biggest conceptual challenges for students (Emden et al., 2018), which justifies significant efforts being devoted to this stage in chemistry education. Later, students should move towards the conceptualisation of a substance as composed of particles with similar shape and size that are not further divisible into smaller particles, and an understanding that a substance is characterised by specific attributes (such as a particular composition or structure) (Talanquer, 2009; Stevens et al., 2010). At this stage, it is common to introduce students to the Periodic Table (Stevens et al., 2010) and its building principle (Emden et al., 2018) as a way to reinforce students’ development of their ideas about the particle model of matter. Research suggests that at this point the periodic table should only be presented as the periodic table of the elemental substances, that is in relation to different macroscopic phenomena (Emden et al., 2018; Traver, 2018), and focusing on the approximately 20 most common elements (Stevens et al., 2010). However, a general search on the Internet will demonstrate that most of the available educational resources for middle or lower secondary school deal with teaching the periodic table of the elements, therefore including information about the subatomic structure of matter. This situation reflects the educational approach of some teachers to “prepare” students in early grades of secondary education for the further challenge of mastering the periodic table in later grades.

Introducing the Periodic Table and its trends creates a need for students to develop a basic model of the atomic structure of matter (Emden et al., 2018; Stevens et al., 2010), where protons, neutrons, and electronic configuration are studied. In middle and high school, a solar system model of the atom (or Bohr model) is usually enough to develop the understanding of the most relevant ideas about elements and the periodic table, and might even be more useful than the electron cloud model (Stevens et al., 2010). For example, the idea of electrons being distributed in shells around the nucleus, with only a certain number of electrons allowed in each shell, connects to the idea of valence electrons and states of oxidation. However, an accurate derivation of the periodic table based on the subatomic properties of elements (e.g. ionisation-energy graphs), which would be ideal to understand its forming principles, is usually too complex to achieve in the curriculum's usual time parameters because it requires a significant amount of time (Bierenstiel and Snow, 2019, p. 1368). For this reason, this inquiry-based approach contrasts with the traditional approach to teaching the periodic table, often based on lecture-style instruction and memorisation of the elements and their properties over several grades, for example by using engaging resources to increase the recall capacity of students, such as crossword puzzles or bingos (Bierenstiel and Snow, 2019).

Sometimes, in upper secondary education, some ideas of the quantum mechanical model of atomic structure (e.g. wave/particle duality, uncertainty, quantisation of energy levels and spin) are introduced qualitatively (Stevens et al., 2010). However, these ideas are difficult for students to conceptualise. In fact, research shows that, even after a year of high school chemistry, a significant portion of students will still hold ideas related to the particle model of the atom (Stevens et al., 2010). This explains why submicroscopic explanations of chemistry are a major area of concern for teachers, as they feel that it is a great challenge to develop the desired ideas of the structure of matter in their students after they complete compulsory education. For this reason, teachers might tend to introduce these ideas early in their teaching sequences so that students can be familiarised with them early and have more opportunities to revisit them. However, addressing the subatomic model too quickly usually leads to a divorce from the macroscopic level, leaving students to accept what teachers tell them and being able to do little more than manipulate chemical ciphers (Nelson, 2002). This dilemma is even more complex if, within this debate, we consider how students should develop formal models of communication in chemistry, such as chemical nomenclature and formulae in secondary school.

Learning chemistry also involves adopting the systems of symbols and representations commonly used within the chemical community (Taber, 2013). Unfortunately, although most curricula propose a stepwise, recursive development of the main scientific ideas, they fail to propose the same developmental perspective for representations (Taskin and Bernholt, 2014, p. 177). Within this lack of guidelines, the introduction to chemical formulae normally starts by linking symbols and formulae to the respective names of the elements and compounds (Taskin and Bernholt, 2014). Rules of chemical nomenclature are frequently introduced in the first few years of compulsory secondary school, and are revisited and expanded every year, without students knowing or having seen the compounds they formulate (Duncan and Rivet, 2013). Therefore, most resources used in school chemistry lessons and even at university level are based on retrieval skills, such as the betting game on nomenclature of the main functional groups (da Silva Júnior et al., 2018) or bingo games (Bayir, 2014). There is not an easy remedy to help students to understand chemical formulae more comprehensively. On top of the conceptual difficulties involved in understanding the structure of matter at different levels, they have to face difficulties related to the symbolic level, such as problems with the syntax of chemical formulae, misunderstanding or difficulties in deciding about when to assign which meaning to a chemical representation (at a structure level, or a reactive level) or when to select a specific representation for a particular purpose (Taskin and Bernholt, 2014). Moreover, there is no consensus in the literature about the adequacy of introducing chemical formulae in secondary school. Some authors, such as De Jong and Taber (2015), argue for introducing chemical nomenclature and formulae from the very beginning, as students’ ideas develop from the particle to the atomic particle models; other authors (Duncan and Rivet, 2013; Taskin and Bernholt, 2014) suggest that chemical formulae should only be introduced when students have already developed some ideas related to the subatomic model of the structure of matter, as the symbolic language should be linked both to particles and processes at the submicroscopic level and to the observable phenomena. Actually, this approach is the one most frequently adopted, as nomenclature is most widely introduced in upper secondary school in teaching programmes (Chi et al., 2018). In addition, and independently of when chemical nomenclature is introduced, Johnstone (2000) and Taber (2013) argue for a simplification of the formulae used, reducing the load of the symbolic representations in order to facilitate students’ development of the desired ideas about the structure of matter.

Regarding the teacher and the role of assessment

The vision of the purpose of assessment is the third main factor heavily conditioned by teachers’ pedagogical beliefs, as described in previous studies such as Lupión-Cobos et al. (2017). The purpose of the assessment, in turn, influences the way students learn, as it plays a fundamental role in educational systems and yet can also be a gateway of change (Wyatt-Smith et al., 2014). It has been believed that in traditional summative assessment, such as exams and marks, students are expected to show that they know how to respond correctly to academic and reproductive questions (i.e., reproduce skills or information demonstrated or provided for students by the teacher) related to what they have studied in the classroom (Curtner-Smith et al., 2001, p. 181). In contrast, from a competence-based learning perspective, assessment-even if it is summative-plays a primarily formative purpose, geared toward identifying mistakes and difficulties throughout the learning process and overcoming them (Wiliam, 2011). From this perspective, assessment seeks to find out whether students know how to apply (transfer) the knowledge they learned in new contexts and real situations, and therefore the questions are productive, i.e., students produce skills or information with which they were formerly unfamiliar (Curtner-Smith et al., 2001, p. 181).

The difference between these visions is not trivial, as it conditions not only the questions in exams but also those asked during learning (Chin and Osborne, 2008). Reproductive questions only allow a formal, closed-ended answer which entails identifying and defining, as the student should have memorised the content shortly before the exam in order to provide the correct answer or perform calculations in exercises similar to ones practised previously. In contrast, productive questions are usually open-ended and complex, and they entail having the student demonstrate that they are capable of reasoning in the response by interrelating a range of ideas and data which have been incorporated into their memory during the learning process (Bransford et al., 2000). Hence, productive questions are intrinsically related to the competence-based educational approach.

Although educational guidelines of all countries currently tend to promote productive assessment in relation to competence-based learning, and those external assessments administered–such as PISA exams even reinforce it, studies nonetheless show that instructors’ teaching approaches and the way they assess their students is not changing (Murchan, 2018). This difficulty in transforming teaching practices which, as previously said, is related to their pedagogical beliefs, can create different educational approaches to science coexisting within the same school, jeopardizing the curricular consistency of an educational centre. However, would this diversity in teaching approaches compromise students’ learning achievement?

Research problem scenario: could competence-based chemistry teaching in grade eight harm students’ performance in grade nine traditional exams?

As stated above, even though the latest curricular reforms in Catalonia (Spain) focus on competence-based teaching and learning and a more formative view of assessment, the way this paradigm materialises in secondary schools (12–16 year-old students) is quite diverse because of the different pedagogical beliefs of teachers. Even if we do not agree with it, unfortunately some reluctant voices believe that competence-based instruction in the early grades of secondary education might harm students’ performance in upper high school traditional exams because they do not arrive “well prepared” (Abril Gallego et al., 2014).

This same debate took place in an urban public secondary school in the Barcelona metropolitan area. In this school students mainly came from middle-class families, and the school had been historically recognised for its academic rigour and excellence together with transmissive instruction. In the prevailing school culture (school ethos), compulsory secondary education was conceived as a “preparation” for post-compulsory secondary education stages. For this reason, chemistry instruction was mostly based on reproducing propaedeutic contents from textbooks and preparing exams through rote learning. In particular, compulsory chemistry education in the school included two courses: introduction to chemistry in grade eight, and intermediate chemistry in grade nine. At the end of grade nine students were expected to master the periodic table and chemistry nomenclature, on the assumption that this was a previous step for non-compulsory advanced chemistry in grades ten, eleven and twelve, following the educational approach described in Abril Gallego et al. (2014). To this end, the exams in grade nine were strongly oriented towards memorising the periodic table and formulating chemistry compounds, always under the assumption that this was good preparation for advanced courses. Some examples of exam questions can be found in the Appendix.

In this context, at the beginning of academic year 2018–19, the grade eight chemistry teaching staff were engaged in a debate about how to define the learning goals in grade eight. It was well known that those grade eight students would later have to face traditional exams in grade nine, but should grade eight be understood as a preparation for grade nine exams or not? Two teachers (hereinafter they are named teachers A and B) were in charge of the instruction in grade eight. They had similar ages (thirties) and similar teaching experience but very different teacher identities regarding the elements discussed above (their role as teachers, the role of students, the role of the content and the role of assessment). These divergences were translated into two divergent views about how to define the learning goals in grade eight:

• Teacher A considered that grade eight had to be a preparation of grade nine, and therefore students should start memorising the periodic table and mastering the formulation rules as soon as possible, in order to be better prepared. In this approach, a competence-based approach proposed by the curricular reform in Catalonia would harm their performance in advanced courses, because it would be “a waste of time”.

• Teacher B advocated implementing a competence-based instructional approach aligned with the curricular reform explained above, engaging students in inquiry and modelling about the particulate model of matter without introducing the periodic table or the chemistry formulation rules.

These divergent views were translated into two different teaching approaches (Table 1). Teacher A followed the pedagogical approach used so far in the school: a transmissive or teacher-centred approach, focused on the definition of chemical concepts, memorisation of the periodic table, work out mathematised reproductive chemistry exercises and procedures, such as those needed for chemical nomenclature, the conversion of units with the factor-label methods, the calculation of solubility, etc. As stated above, teacher A considered that those concepts and skills were a preparation for grade nine, and thus her exams were focused on reproductive demands, far from the competence-based approach discussed above.

Table 1 Example of activities developed by teachers A and B. More details of the teaching approach by teacher B are provided in López-Simó and Garrido-Espeja (2021)

Common features of the teaching programme given by Teachers A and B in grade eight

• Lectures, discussions, and classroom activities about basic chemistry: laboratory equipment and its functions, laboratory safety rules, proprieties of matter (mass, volume, density, temperature…), particulate model of matter to explain those proprieties, changes of state and temperature – time graphics (melting and boiling point), typologies of mixtures and methods of separating mixtures, solubility and precipitation, acids and bases, introduction to pH scale.

• Laboratory practices: one hour per week students from both groups attended a lab for practical activities related to the topics mentioned above (example: measurement of density; separation of homogeneous and heterogeneous mixture techniques; measurement of temperature at boiling point, etc.).

• Outdoor and special activities: students from both groups participated in a visit to the local purification plant and in different crosscutting projects, such as a medieval project where students had to create ink with traditional techniques.

---

Specificities of the approach used by Teacher A	Specificities of the approach used by Teacher B
Expanded programme: beside the common programme, teacher A included propaedeutic content for grade nine:	Shorter programme: teacher B just focused on the common programme, avoiding this propaedeutic content, but dedicating more time and more attention to the basic topics presented above.
• Conversion of units with factor-label method
• Structure of the atom: sub-atomic particles, ions, and isotopes.	This allowed a more competence-based approach to be provided, including activities such as:
• Definition and structure of the periodic table, including memorisation of the position of the most relevant elements.	• Dialogic discussions to connect abstract concepts to everyday life phenomena: weather, smelling, meals, floating and sinking, etc.
• Introduction of the chemical nomenclature, including hydride, oxides, and binary salts.	• Inquiry about changes of state in matter, dedicating more time to defining research questions, making predictions, designing experiments, analysing results, etc.
Reproductive approach: teacher A placed more emphasis on reproductive demands:	• Modelling the particulate model of matter with clay and drawings.
• Reading the textbook.	• Using virtual simulations to investigate the properties of matter, connecting the microscopic and macroscopic representation.
• Definition of concepts.
• “Fill the gap” exercises.
• Reproductive mathematical calculations.

---

Conversely, teacher B incorporated a series of changes in the teaching approach. These changes entailed modifications in the content, teaching methodology and assessment with the goal of promoting productive and more competence-based instruction to develop their ideas about the particle model of matter. Therefore, teacher B did not introduce any idea about the subatomic model or chemical nomenclature in grade eight; nor did they focus on these mathematised exercises and procedures. In contrast, more time and effort were dedicated to understanding simple physics and chemistry phenomena by using the particulate model of matter, such as conceptually understanding the changes of state, the mixtures and diffusion, the light-matter interaction, and also the nature of science. These changes did not, however, only concern the contents to be taught, but also the methods. Teacher B gave students a more active role and offered a more productive and formative assessment by engaging them in modelling and inquiry practices.

From among the total of 138 students enrolled in the grade eight chemistry course, 78 of them received the instruction from teacher A (three class groups consisting of 25, 26 and 27 students, respectively), and 54 students received instruction from teacher B (two class groups consisting of 27 each). Henceforth, these students will be named groups A and B. We must remark that these two groups were equivalent according to composition, since in the school groups are defined according to homogeneity criteria, to avoid any kind of segregation (gender, race, social background, interest in science, academic record, etc.).

Research questions

Right before the start of grade nine in the following academic year (2019–2020), these five class groups were reorganised to form new groups (see Fig. 1) as a pedagogical means of promoting personal relations among students within the school, but also assuring the homogeneity discussed above. In grade nine there were no differences between teaching approaches between class groups. Therefore, both groups A and B from the 2018–2019 academic year received the same transmissive approach corresponding to group A in 2019–2020, in coherence with the school ethos. This situation entailed instructional programme coherence for group A but not for group B, which experienced major changes in the educational demands they were expected to meet in the subsequent academic year.

Fig. 1 Diagram of the organization of the groups A and B within grade eight and grade nine, and the different data-collection instruments used over time.

Based on the characteristics of the context, the following research questions are targeted towards understanding the potential impact of the different teaching approaches in the subsequent academic year by means of comparing students’ performance, in coherence with the school ethos in which the research takes place:

• Q1 – Are there differences between students in group A and group B in their performance in an assessment test about the structure of matter and understanding of chemical formulae at the beginning of grade nine?

• Q2 – Are there differences between students in group A and group B in their performance in exams on ideas about the structure of matter and understanding of chemical formulae throughout grade nine?

• Q3 – Are there differences in students’ final evaluation at the end of grade nine between students in group A and group B?

Method

This research involves an exploratory case study in chemistry education in a secondary school in Catalonia (Spain), following Yin (2014). Hence, its aim is to generate generalisable theoretical propositions from the analysis of the case, and not from the population (Yin, 2014, p. 15). Particularly, it seeks to discover whether an instruction with a competence-based approach (student-centred, promoting the construction of the ideas about the particle model of matter, and based on formative assessment) which was being implemented in some of the class groups of the second year of secondary school (grade eight) may have a negative impact on students’ development of their ideas about the structure of matter in the next year (grade nine), when all students would be taught with a transmissive approach (involving a high level of abstraction and complexity, and based on rote learning of the subatomic model of matter and summative assessment). Following the school policy, when moving from grade eight to grade nine, new class groups were made, randomly mixing students from previous groups, as described in the following subsection. This measure contributed to reducing the bias of the group effect and increasing the internal validity (Cohen et al., 2018).

In this research, data collection instruments were presented to students in Catalan, as it was the language of instruction, and students used this language to provide their answers. The authors analysed data in the original language for enhancing the reliability of the research, as it is also their primary language of use. In this article, data collection instruments and data extracts have been translated for a better understanding, although assuming that some nuances of the original files may be lost (Taber, 2018). Translations were initially performed by each author and contrasted to reach a consensus, revised afterwards by the university official proof-readers, and finally assessed again by the authors for reliability, following the guidelines on Taber (2018).

Research design and data-collection instruments

Data to answer the three research questions were collected from students’ summative evaluation results throughout grade nine, following a similar design to Koksal and Berberoglu (2014). All the students took part in five different tests: an initial assessment (E0) and four exams (E1, E2, E3 and E4) that covered all content related to the nature of matter from grade nine, according to the Catalan curriculum. Plus, each student received a mark at the end of the academic year in their school report (FA), as shown in Fig. 1. The grade nine teacher designed all assessments following the demands of the Catalan curriculum, except for the initial assessment (E0) which was co-designed with grade nine and grade eight teachers.

The initial assessment (E0) was designed specifically to analyse what groups A and B remembered from the previous year and was administered at the beginning of the academic year 2019–20 (Fig. 1). The test included six questions, three reproductive and three productive, which either both groups or only one group had studied in grade eight, as shown in Table 2. Questions E0.2 and E0.3 asked for a retrieval of propaedeutic content only taught to group A with reproductive questions, and E0.5 asked for a productive explanation of the content only taught to group B. Groups A and B were similarly prepared for questions E0.1, E0.4 and E0.6. Hence, there was one more specific question in which just group A had previous experience. As the amount of content covered in group A had been broader, it was agreed to gather more information about their previous knowledge as a way to better characterise their learning and, therefore, increase the validity of results.

Table 2 Questions included in the initial test and their type, as well as each group's previous experience, according to the type of instruction they received in grade eight

Name and description of the question	Type of question	Previous experience Group A	Previous experience Group B
E0.1. Nomenclature of laboratory instruments.	Reproductive	Yes	Yes
E0.2. Exercise in converting units of measurement (conversion factors).	Reproductive	Yes	No
E0.3. Formulae of binary compounds.	Reproductive	Yes	No
E0.4. Explanation of a separation of compounds.	Productive	Yes	Yes
E0.5. Explanation of a floatation phenomenon.	Productive	No	Yes
E0.6. Explanation of a phenomenon involving a change of state.	Productive	No	No

---

During grade nine, all students took four exams (E1–E4), as shown in Table 3. This assessment was monitored by their grade nine chemistry teacher, a different individual from teachers A and B in grade eight. Exams E1–E4 had a reproductive and rote-learning style, in coherence with the prevailing school culture, as exemplified in Fig. 1. It is worth noting that in grade eight group A had studied considerably more of the content of exams E1–E4 than group B. Because of the lockdown imposed in Catalonia from March to May 2020 due to the COVID19 pandemic, all the teaching activities and evaluations E3 and E4 were done online. Grade nine chemistry teachers took actions to avoid cheating and plagiarism in those online exams, such as using web cams and specific software.

Table 3 List of exams related to the main content taught. Detailed examples are presented in the Appendix

Exam	Content already known by A and B	Content already known by A but new for B	Content new for both A and B
E1	Physical properties of matter.	International system of units.
E2	Pure and mixed substances.	Conversion factors between units.	Solutions, concentration, and molarity.
E3 (online)	Corpuscular theory.	Atomic structure and periodic table.	History of atomic models; Isotopes.
E4 (online)		Chemical nomenclature of oxides and hydrides.	Chemical nomenclature of binary salts and ions.

---

As the items of the E1–E4 evaluations were already used in the previous academic years, no pilot study was conducted. The content validity of E0–E4 evaluations was assessed through expert opinion (i.e., researchers and other teachers). The congruence between each item content was investigated, as were the content covered in each exam, its relationship with what Group A and Group B had been taught in the previous years, and the objectives of the instructional programme, following Koksal and Berberoglu (2014).

Ethical considerations

This research was conducted in accordance with the principles embodied in the Declaration of Helsinki, the British Educational Research Association (2018) Ethical Guidelines for Educational Research and in accordance with local statutory requirements. Participant teachers and students provided informed consent to participate voluntarily, and were treated fairly, sensitively, and with dignity. The school principal also provided informed consent and authorisation to conduct the research. Among other measures, confidential and anonymous processing of participants’ data was ensured.

Data analysis

As a first stage, a qualitative analytical approach was undertaken to obtain students’ results from each evaluation. Particularly, in E0, each answer was coded independently, as follows:

• L1: blank (no answer) or answers unrelated to the question.

• L2: answers that include some correct parts but that were overall incorrect, tautological, or redundant with the question.

• L3: partly correct answers, but with some ambiguities or erroneous aspects.

• L4: answers that were correct in both form and content.

Table 4 shows one example of categorisation of the answer to question E0.4, where students were asked about the separation of a heterogeneous mixture of sand, salt, and iron. In E0.4, the answer coded as L2 explains a mixture separation method that does not serve the suggested challenge. The answer coded as L3 suggests a proper method (crystallisation and magnetic separation), but wrongly states that “salt evaporates with water”. The answer coded as L4 describes a process composed of three separation methods in the correct order.

Table 4 Examples of answers for E0.4 and their coding

	E0.4. In a waste treatment plant, they need to separate a mixture of sand, salt, and ferrous materials. Explain the steps that they should follow to perform this separation and why.
L2
	Translation: we need to have some strainers where each one has smaller and smaller holes and thus be able to separate each mixture
L3
	Translation: i would put water in it and heat it so that the salt would evaporate with the water. Once the salt is separated, just pass a magnet over the sand with ferrous materials. As the magnet attracts the iron, we would get the result we expect.
L4
	Translation: we pass a magnet over it and the ferrous materials will stick. We put the mixture in water and mix, the salt will dissolve. We pour the sand in water through the filter paper and the sand will stay there. Then we distil the water with the salt.

---

For E1–E4 evaluations, the teacher gave a mark for each student's exam using numerical values from 0 to 10, according to the school grading system. Finally, we also collected the final chemistry course assessment marks (FA) for each student. According to the evaluation policy in Catalonia, these marks had to be expressed under four categories: Not Attained (NA), Satisfactory Attainment (SA), Good Attainment (GA) and Excellent Attainment (EA). However, in actual fact, according to the prevailing school culture, the teacher gave a numerical equivalence, in which NA corresponds to the range of marks 0–4.9, SA 5–6.9, GA 7–8.4, and EA 8.5–10. All students’ evaluations were carried out by the grade nine teacher, which was considered to provide internal validity and reliability of results, following Cohen et al. (2018).

The diverse range of tests administered generated a database with a combination of categorical and numerical variables, which was analysed in a second stage of results’ analysis. The database was read and managed with the program JASP 0.13. Univariate frequency tables of tests E0 and FA were prepared as well as tables with the summary of the main statistical values for the continuous variables (E1, E2, E3 and E4). As a first step, normality of the data was assessed in the continuous variables (E1–E4) using the Shapiro-Wilk test, and following Khusainova et al. (2016). Only variable E2 did not deviate significantly from the normal distribution (W = 0.982, p = 0.077), and there were no significant differences between the variances of the two groups (A and B) in the Levene test (F(1) = 1.186, p = 0.278).

The contrasts to identify differences between groups A and B in the different assessments were conducted via the Mann-Whitney U-test for independent samples in exams E1, E3 and E4, following (Cohen et al., 2018). This same test was used for the ordinal variables of the initial and final exams (E0 and FA) due to the nature of the data. The data in exam E2 were contrasted using the Student t-test for independent samples, given that this variable had shown a suitable fit with the normal distribution.

A two-tailed test was used in the contrast for the initial assessment (E0), given that the sign of the potential difference between the two compared medians was unknown, following Hazra and Gogtay (2016). However, one-tailed tests were used for the contrasts between E1, E2, E3, E4 and FA, given that the alternative hypothesis to be checked was whether the median in group A was higher than the median in group B. Additionally, to complement this, the Chi-squared test was used to check for possible differences by each range of marks in the final exam (FA). A confidence interval of 95% was defined for all the contrasts.

Results

Overview of the results

Fig. 2 shows the frequencies found in the different questions of the initial assessment (E0) by the students in group A and group B. Some visual differences between scores of groups A and B can be seen in questions E0.2, E0.3 and E0.5, while results in E0.1, E0.4 and E0.6 appear to be similar. Fig. 3 displays the mean of the results for assessments E1, E2, E3 and E4 of groups A and B, as well as the associated standard error. This figure demonstrates greater differences between the two groups in the results for assessments E2 and E4. Finally, Fig. 4 shows the frequencies for groups A and B in the final assessment (FA), where final grades of students of both groups appear to be visually similar.

Fig. 2 Proportion and frequencies for each level (L1–L4) and question on the initial assessment (E0) in groups A and B (N_A = 78; N_B = 54).

Fig. 3 Means of students’ grades in assessments E1, E2, E3 and E4 for groups A and B and standard error bars (N_A = 78; N_B = 54).

Fig. 4 Frequencies and percentages for each level of attainment in the final assessment for group A and group B (N_A = 78, N_B = 54).

Differences between groups A and B

Results of the contrasts between group A and group B in the initial assessment, in Table 5, reveal statistically significant differences between groups A and B for questions E0.2, E0.3 (reproductive) and E0.5 (productive), considering a significance level of p < 0.05. The positive correlation coefficient for questions E0.2 and E0.3 shows that the mean of group A is significantly higher than group B. In contrast, the effect size is small and moderate in questions E0.2 and E0.3, respectively. Hence, the differences between the groups in these questions are small, albeit statistically significant. In question E0.5, the negative correlation coefficient shows that the mean of group B is significantly higher than group A. We should add that, just like with question E0.3, the magnitude of the effect is moderate.

Table 5 Results of the contrasts between the results of groups A and B in assessments E0, E1, E2, E3, E4 and FA (N_A = 78, N_B = 54)

Question or assessment	Statistical value	Degrees of freedom	p value	Effect size^a
a For E0, E1, E3, E4 and FA, the effect size is calculated by the rank-biserial correlation coefficient. For E2, the effect size is reported via Cohen's D.
E0.1	U = 2085.000	—	0.266	0.103
E0.2	U = 2257.500	—	0.033	0.194
E0.3	U = 2725.500	—	<0.001	0.442
E0.4	U = 1974.500	—	0.656	0.045
E0.5	U = 1053.000	—	<0.001	-0.443
E0.6	U = 2136.000	—	0.176	0.130
E1	U = 2229.000	—	0.285	0.058
E2	t = 1.764	129	0.040	0.314
E3	U = 2252.000	—	0.250	0.069
E4	U = 2347.500	—	0.065	0.158
FA	U = 2155.500	—	0.329	0.043

---

Regarding the results of assessments E1, E2, E3 and E4, a statistically significant difference was only found between the mean of group A and group B in assessment E2, with a higher mean in group A. However, the effect size is small (Table 5). The contrasts do not reveal statistically significant differences between the means of groups A and B in the final assessment (FA) for a confidence interval of 95%. Furthermore, statistically significant differences were not found in the FA in the number of students in the two groups with excellent attainment, X² (1, N = 17) = 2.882, p = 0.090; good attainment X² (1, N = 27) = 0.037, p = 0.847; satisfactory attainment X² (1, N = 64) = 3.063, p = 0.080; or no attainment X² (1, N = 24) = 0.167, p = 0.683.

Conclusions

During academic year 2018–19, the grade eight students in group A (n = 78) and group B (n = 54) received considerably different educational approaches in their chemistry lessons (Table 1). We were aware of the benefits that competency-based teaching instruction focusing on the development of the particle model of matter might have over students’ learning, but also wondered about the risks of an isolated instructional approach within the traditional school ethos. To this end, we investigated what possible influence this educational shift might have had on students’ performance in the following year (grade nine), where chemistry lessons were focused on developing students’ ideas about the subatomic model of the structure of matter.

To answer research question Q1 we coded and contrasted students’ responses in initial test E0 (Table 3), comprising three reproductive questions (E0.1, E0.2 and E0.3) and three productive questions (E0.4, E0.5 and E0.6). We found significant differences between the two groups for the responses to E0.2 and E0.3 (both reproductive) and E0.5 (productive) (Table 5). With E0.2 and E0.3, we found that group A scored significantly higher than group B. However, we should note that although these differences are statistically significant, the effect size is small for E0.2 and moderate for E0.3, given that more than half the students in group A did not answer question E0.2 and 40% did not answer question E0.3. These results show that preparatory instruction on the conversion factors and binary formulation might have had some influence, as at the beginning of the next academic year students from group A could perform better than students from group B, who had no prior experience on these topics. With E0.1, we found that both groups of students had similar difficulties remembering the name of some of the laboratory instruments they all had to memorise the previous year, as most students in both groups did not answer this question correctly.

Regarding the productive questions, we found that results for E0.4 were similar in both groups (they both had received previous experience about the topic) and were also much higher than the results of the three previous questions, especially E0.1 (Table 5). One year later, most students were much more capable of explaining how to use laboratory instruments to separate a mix than remembering the names of the instruments used. These results concur with results of previous studies, in which productive demands -in which students have to come up with an explanation and not just reproduce a definition, name or procedure-allow students to activate more knowledge in the medium term (Bransford et al., 2000). In question E0.5 we see the opposite, as in E0.2 and E0.3, as group B (who had learned how to explain objects floating and sinking, a productive demand) scored significantly higher than A, which had no previous experience in this. The effect size is moderate, with a magnitude virtually identical to that of E0.3. Finally, the answers to E0.6, which neither of the two groups had covered before, showed similar results, and no significant differences were found.

In short, we found differences in the scores of the answers to questions E0.2, E0.3 and E0.5, which were coherent with the type of chemistry educational approach been taught to each of the two groups (Table 1). This first group of results enables us to draw two conclusions. First, there is a suggested relationship between the instruction received in grade eight and the performance in the initial test in grade nine. We found differences in both the contents that were only taught in group A (E0.2 and E0.3) and those only taught in group B (E0.5), while these differences disappear in the content taught to both groups (E0.1 and E0.4) or in content that is new for both groups (E0.6). Second, we found that for the questions in which students had previous experience, those that entailed a reproductive demand (E0.1, E0.2, and E0.3) earned worse scores than those that entailed a productive demand (E0.4, E0.5).

Regarding research question Q2, results in Fig. 3 show that of the four exams taken during grade nine (E1–E4), the mean scores of group B were slightly lower than those of group A, although significant differences were only found in E2 (Table 5). In this exam, students from group A scored significantly higher on average than those from group B, in coherence with the differences observed in question E0.2, as E2 also included exercises on conversion factors. In contrast, even though group A in grade eight was taught content related to the subatomic model of the structure of matter, such as atomic structure, the periodic table (assessed in E3) and the formulation of oxides and hydrides (assessed in E4), students in group B overall achieved similar results or, at least, there were no significant differences between the two groups. This would suggest a scarce effect of the instruction of content about the subatomic model in grade eight to prepare students for more challenging and future learning demands, as Talanquer (2018) discusses in his study. Consequently, our study points to the need to take the opportunity to study in greater depth the ideas about the particle model of matter before moving to a more complex understanding of matter, as Talanquer (2009), Taber and García-Franco (2010), Tsitsipis et al. (2012) and other authors suggest. Students who had not memorised the period table (E3) or learned formulas (E4) in grade eight achieved similar results to students who had, although students from group B were at a slight disadvantage because they had not mastered the conversion factor units (E2). Furthermore, by examining the size of the standard deviation bars in Fig. 3, we can see that there were many variations in the results within each group in the four intermediate tests (that is, a broad distribution of scores). From our perspective, this result cannot be explained only by the different instruction received in grade eight but also by many other personal and social factors of the students, as extensively discussed in the literature, such as in Morell et al. (2017), and Zuber and Altrichter (2018). With the information available to us, and accepting the limitation of the sample used, in our research we argue that it is dubious whether the preparatory instruction that group A received in grade eight facilitated its performance in grade nine, but that more research should be done to shed light on this topic.

Finally, regarding research question Q3, Fig. 4 shows a similar distribution of final assessments (FA) between groups A and B, where no statistical differences in the overall distribution of results were found. That is, the distribution of students with different attainment levels on their school report is similar in groups A and B. When analysing the chart in Fig. 4 and focusing on the students that aspired to obtain a higher mark (EA), we do find a slight visual difference between the two groups, but the proportion of students with good and excellent attainment marks are exactly the same.

Overall, results from our research suggest that we cannot establish a strong relationship between having received a propaedeutic teaching approach, in which content about the atomic structure, periodic table and formulation are introduced early, and a positive effect on students’ performance in the subsequent academic year, unlike what some teachers believe. Our results question previous findings in which students following instructional programme coherence outperformed their peers (Shin et al., 2019), suggesting a need to review the influence of other factors in students’ outcomes. Therefore, a competence-based approach, though not as preparatory for “hard” material or coherent, did not significantly diminish students’ overall performance in the following year.

In this study, we were not able to measure other possible effects of the teaching in group B, such as their possible developed competence in constructing explanations using both models of matter, and in solving problems from everyday life using chemical reasoning, as the assessment in grade nine was eminently reproductive. Moreover, the influence of psychological and emotional factors in learning outcomes, such as motivation, self-efficacy and ownership, were not considered. In addition, we must consider that in March 2020 there was a lockdown that led teachers to transfer face-to-face instruction to an online method. Without undermining the important effect of these factors on students’ outcomes, we attribute our findings to the differences in the instructional programmes of grade eight, since both Group A and Group B were mixed into homogeneous groups in grade nine, and the shift from face-to-face to online instruction affected all students equally. Finally, we were unable to implement more sophisticated research methods, such as the involvement of both grade 8 teachers in both approaches, due to the constraints of the research scenario.

Despite the limitations of our research design and sample, we believe that the fact that no negative effect on group B was found can stand as an important argument in favour of the shift towards more competence-based educational approaches, providing more opportunities to students to expand their ideas about the structure of matter. Students need time to make sense of the macroscopic and microscopic concepts in order to be fluent in the technical terminology used to communicate them (Taber, 2013). The fear of a lack of future preparation (which leads some teachers to turn to the preparatory logic) should not curb the shift towards more competence-based approaches. On the contrary, lowering the number of contents to make them more meaningful, robust, and closely tied to students’ reality (in short, more competence-based) should be viewed as an opportunity to develop their scientific literacy not just for the next year at school but for their entire lives.

Conflicts of interest

All authors declare that there are no potential financial or non-financial conflicts of interest.

Appendix

Example of different reproductive exercises and questions included in the different evaluation tools in grade nine, translated from the original (Catalan). Most of the exercises of E1, E2, E3 and E4 do not involve understanding specific models in chemistry to develop an explanation about chemistry phenomena, but simply the use of different rules to reach the solution.

Questions included in E0

Example of question corresponding to E1

Example of question corresponding to E2

Example of question corresponding to E3

Example of question corresponding to E4

Acknowledgements

Research financed by the Ministry of Science, Innovation and Universities (PGC2018-096581-B-C21) and the ACELEC (2021 SGR 00647) and ARGET (2017SGR1682) research groups.

References

1. Abril Gallego A. M., Romero Ariza M., Quesada Armenteros A. and García, F. J., (2014). Creencias del profesorado en ejercicio y en formación sobre el aprendizaje por investigación, Revista Eureka Sobre Enseñanza y Divulgación de Las Ciencias, 11(1), 22–33 DOI:10.25267/Rev_Eureka_ensen_divulg_cienc.2014.v11.i1.04.
2. Bayir E., (2014), Developing and playing chemistry games to learn about elements, compounds, and the periodic table: Elemental Periodica, Compoundica, and Groupica, J. Chem. Educ., 91(4), 531–535 DOI:10.1021/ed4002249.
3. Bierenstiel M. and Snow K., (2019), Periodic Universe: A Teaching Model for Understanding the Periodic Table of the Elements, J. Chem. Educ., 96(7), 1367–1376 DOI:10.1021/acs.jchemed.8b00740.
4. Biesta G., Priestley M. and Robinson S., (2015), The role of beliefs in teacher agency, Teachers and Teaching: Theory and Practice, 21(6), 624–640 DOI:10.1080/13540602.2015.1044325.
5. Bransford J. D., Brown A. L. and Cocking R. R., (2000), How People Learn: Brain, Mind, Experience, and School: Expanded Edition, National Academy Press.
6. British Educational Research Association (BERA), (2018), Ethical Guidelines for Educational Research, 4th edn, https://www.bera.ac.uk/researchers-resources/publications/ethical-guidelines-for-educational-research-2018.
7. Cheung D. and Ng P.-H., (2000), Science teachers’ beliefs about curriculum design, Res. Sci. Educ., 30(4), 357–375 DOI:10.1007/BF02461556.
8. Chi S., Wang Z., Luo M., Yang Y. and Huang M., (2018), Student progression on chemical symbol representation abilities at different grade levels (Grades 10–12) across gender, Chem. Educ. Res. Pract., 19(4), 1055–1064 10.1039/c8rp00010g.
9. Chin C. and Osborne J., (2008), Students’ questions: a potential resource for teaching and learning science, Stud. Sci. Educ., 44(1), 1–39 DOI:10.1080/03057260701828101.
10. Cohen L., Manion L. and Morrison K., (2018), Research Methods in Education, 8th edn, Routledge.
11. Cokelez A., (2012), Junior High School Students’ Ideas about the Shape and Size of the Atom, Res. Sci. Educ., 42(4), 673–686 DOI:10.1007/s11165-011-9223-8.
12. Curtner-Smith M. D., Todorovich J. R., McCaughtry N. A. and Lacon S. A., (2001), Urban Teachers’ Use of Productive and Reproductive Teaching Styles Within the Confines of the National Curriculum for Physical Education, Eur. Phys. Educ. Rev., 7(2), 177–190 DOI:10.1177/1356336X010072005.
13. da Silva Júnior J. N., Sousa Lima M. A., Nunes Miranda F., Melo Leite Junior A. J., Alexandre F. S. O., de Oliveira Assis D. C. and Nobre D. J., (2018), Nomenclature Bets: An Innovative Computer-Based Game To Aid Students in the Study of Nomenclature of Organic Compounds, J. Chem. Educ., 95(11), 2055–2058 DOI:10.1021/acs.jchemed.8b00298.
14. De Jong O. and Taber K. S., (2015), The many faces of high school chemistry, Handbook Res. Sci. Educ., II(JANUARY 2014), 457–480 DOI:10.4324/9780203097267.
15. Duncan R. G. and Rivet A. E., (2013), Science learning progressions, Science, 339(6118), 396–397 DOI:10.1126/science.1228692.
16. Duschl R., Maeng S. and Sezen A., (2011), Learning progressions and teaching sequences: a review and analysis, Stud. Sci. Educ., 47(2), 123–182 DOI:10.1080/03057267.2011.604476.
17. Emden M., Weber K. and Sumfleth E., (2018), Evaluating a learning progression on “Transformation of Matter” on the lower secondary level, Chem. Educ. Res. Pract., 19(4), 1096–1116 10.1039/c8rp00137e.
18. Harlen W., (2015), Working with Big Ideas of Science Education, Science Education Programme (SEP) of IAP.
19. Hazra A. and Gogtay N., (2016), Biostatistics series module 5: determining sample size, Indian J. Dermatol., 61(5), 496–504 DOI:10.4103/0019-5154.190119.
20. Johnson P., (2005), The Development of Children's Concept of a Substance: A Longitudinal Study of Interaction Between Curriculum and Learning, Res. Sci. Educ., 35(1), 41–61 DOI:10.1007/s11165-004-3432-3.
21. Johnstone A. H., (2000), Teaching of chemistry – Logical or psychological? Chem. Educ. Res. Pract., 1(1), 9–15 10.1039/A9RP90001B.
22. Kazempour M., (2009), Impact of Inquiry-Based Professional Development on Core Conceptions and Teaching Practices: A Case Study, Sci. Educ., 18(2), 56–68.
23. Khusainova R. M., Shilova Z. V. and Curteva O. V., (2016), Selection of Appropriate Statistical Methods for Research Results Processing, Int. Electron. J. Math. Educ., 11(1), 303–315 DOI:10.12973/iser.2016.21030a.
24. Koksal E. A. and Berberoglu G., (2014), The Effect of Guided-Inquiry Instruction on 6th Grade Turkish Students’ Achievement, Science Process Skills, and Attitudes Toward Science, Int. J. Sci. Educ., 36(1), 66–78 DOI:10.1080/09500693.2012.721942.
25. López-Simó V. and Garrido-Espeja A., (2021), Substàncies complexes, canvis d’estat i llindar de comprensió del model matèria a 2n d’ESO, Didacticae: Revista de Investigación En Didácticas Específicas, 9, 120–138 DOI:10.1344/did.2021.9.120-138.
26. Luft J. A. and Roehrig G. H., (2007), Capturing Science Teachers’ Epistemological Beliefs: The Development of the Teacher Beliefs Interview, Electron. J. Sci. Educ., 11(2), 38–63, http://ejse.southwestern.edu.
27. Lupión-Cobos T., López-Castilla R. and Blanco-López Á., (2017), What do science teachers think about developing scientific competences through context-based teaching? A case study, Int. J. Sci. Educ., 39(7), 937–963 DOI:10.1080/09500693.2017.1310412.
28. Morell, L., Collier, T., Black, P. and Wilson, M., (2017), A construct-modeling approach to develop a learning progression of how students understand the structure of matter, J. Res. Sci. Teach., 54(8), 1024–1048 DOI:10.1002/tea.21397.
29. Murchan D., (2018), Introducing school-based assessment as part of junior cycle reform in Ireland: a bridge too far? Educ. Assess. Eval. Acc., 30(2), 97–131 DOI:10.1007/s11092-018-9274-8.
30. Nelson P. G., (2002), Teaching Chemistry Progressively: From Substances, to Atoms and Molecules, to Electrons and Nuclei, Chem. Educ. Res. Practice, 3(2), 215–228 10.1039/B2RP90017C.
31. Newmann F. M., Smith B., Allensworth E. and Bryk A. S., (2001), Instructional Program Coherence: What It Is and Why It Should Guide School Improvement Policy, Educ. Eval. Policy Anal., 23(4), 297–321 DOI:10.3102/01623737023004297.
32. OECD, (2019), PISA 2018 Assessment and Analytical Framework, PISA, OECD Publishing DOI:10.1787/b25efab8-en.
33. Sanmartí, N., (2020), Avaluar i aprendre: Un únic procés. Editorial Octaedro, https://octaedro.com/libro/avaluar-i-aprendre-un-unic-proces/.
34. Schleicher A., (2020), TALIS 2018: Insights and Interpretations, in OECD Education Journal, http://www.oecd.org/education/talis/TALIS2018_insights_and_interpretations.pdf.
35. Serin H., (2018), A Comparison of Teacher-Centered and Student-Centered Approaches in Educational Settings, Int. J. Soc. Sci. Educ. Stud., 5(1), 164–167 DOI:10.23918/ijsses.v5i1p164.
36. Shin N., Choi S. Y., Stevens S. Y. and Krajcik J. S., (2019), The Impact of Using Coherent Curriculum on Students’ Understanding of Core Ideas in Chemistry, Int. J. Sci. Math. Educ., 17(2), 295–315 DOI:10.1007/s10763-017-9861-z.
37. Stevens S. Y., Delgado C. and Krajcik J. S., (2010), Developing a hypothetical multi-dimensional learning progression for the nature of matter, J. Res. Sci. Teach., 47(6), 687–715 DOI:10.1002/tea.20324.
38. Taber K. S., (2013), Revisiting the chemistry triplet: drawing upon the nature of chemical knowledge and the psychology of learning to inform chemistry education, Chem. Educ. Res. Pract., 14(2), 156–168 10.1039/c3rp00012e.
39. Taber K. S., (2018), Lost and found in translation: guidelines for reporting research data in an ‘other’ language, Chem. Educ. Res. Pract., 19(3), 646–652 10.1039/C8RP90006J.
40. Taber K. S. and García-Franco A., (2010), Learning Processes in Chemistry: Drawing Upon Cognitive Resources to Learn About the Particulate Structure of Matter, J. Learn. Sci., 19(1), 99–142 DOI:10.1080/10508400903452868.
41. Talanquer V., (2009), On cognitive constraints and learning progressions: the case of “structure of matter”, Int. J. Sci. Educ., 31(15), 2123–2136 DOI:10.1080/09500690802578025.
42. Talanquer V., (2018), Progressions in reasoning about structure-property relationships, Chem. Educ. Res. Pract., 19(4), 998–1009 10.1039/c7rp00187h.
43. Taskin V. and Bernholt S., (2014), Students’ Understanding of Chemical Formulae: a review of empirical research, Int. J. Sci. Educ., 36(1), 157–185 DOI:10.1080/09500693.2012.744492.
44. Traver M. J., (2018), La tabla periódica. Relación entre las propiedades periódicas de las sustancias elementales y las atómicas, Alambique Didàctica de Las Ciencias Experimentales, 93, 47–52.
45. Tsitsipis G., Stamovlasis D. and Papageorgiou G., (2012), A probabilistic model for students’ errors and misconceptions on the structure of matter in relation to three cognitive variables, Int. J. Sci. Math. Educ., 10(4), 777–802 DOI:10.1007/s10763-011-9288-x.
46. Wiliam D., (2011), What is assessment for learning? Stud. Educ. Eval., 37(1), 3–14 DOI:10.1016/j.stueduc.2011.03.001.
47. Wyatt-Smith C., Klenowski V. and Colbert P., (2014), Designing Assessment for Quality Learning, vol. 1, Springer Netherlands DOI:10.1007/978-94-007-5902-2.
48. Yin R. K., (2014), Case Study Research. Design and Methods, 5th edn, SAGE Publications.
49. Zhang L., (2016), Is Inquiry-Based Science Teaching Worth the Effort? Sci. Educ., 25(7–8), 897–915 DOI:10.1007/s11191-016-9856-0.
50. Zuber J. and Altrichter H., (2018), The role of teacher characteristics in an educational standards reform, Educ. Assess., Evalu. Acc., 30(2), 183–205 DOI:10.1007/s11092-018-9275-7.

---