Patterns of Textbook Use in EFL: Adaptation Techniques and Their Effects on Form-Focused and Meaning-Focused Instruction ()
1. Introduction
There is a generalized agreement in the specialized literature about textbooks being the primary pedagogical tool teachers resort to in foreign (L2) language classrooms (Harwood, 2022; Li & Xu, 2020; Rathert & Cabaroğlu, 2021, etc.), despite heated debates about the suitability of textbooks (see Mishan, 2022, for an exhaustive critique of the stances). Scholars also concur that research on textbook use, though slowly growing, is still limited (Graves, 2019; Graves & Garton, 2017; Guerrettaz & Johnston, 2013; Schwab, 2022; Zhang et al., 2022). The imperative need to conduct research in this area is due to the ostensible fact that textbooks “tend to influence what and how teachers teach and, to some extent, how and what students learn” (Zhang et al., 2022, p. 3). Consequently, the role of textbooks in the language classroom is crucially mediated by teachers. They can adopt both planned and spontaneous decisions about the retention of activities or the implementation of adaptations such as the addition and/or omission of certain content or entire activities, modification (restructuring or changing the procedure, reordering activities), etc., with the resulting impact on the learning affordances granted to students.
Also, it is relevant to identify whether and to what extent teachers’ manipulation and/or adherence to textbooks exerts an effect on Form-focused (FFI) and Meaning-focused Instruction (MFI) and, consequently, on the opportunities offered to learners to become satisfactory language users. Given that both types of instruction serve to develop the types of L2 knowledge involved in the attainment of communicative competence (Canale & Swain, 1980; Canale 1983), that is, explicit and implicit (Loewen & Sato, 2017), contemporary classrooms should ensure the coherent integration of both types of instruction. However, research on the presence of FFI and MFI when using textbooks is scant. Therefore, efforts should be directed towards measuring the use of FFI and MFI to determine possible differences (if any) between the corresponding amount of the textbooks’ original activities and those resulting from the teachers’ adaptations. Such information could be valuable to researchers in Instructed Second Language Acquisition (ISLA) undertaking studies aimed to analyze the potential effects of teachers’ textbook use patterns on the development of explicit and implicit knowledge in ecological settings, i.e., real classes where textbooks are the main teaching instrument. Moreover, a systematized procedure agreed upon in the scholarly arena would also facilitate the comparison of results between the different studies. In parallel to these benefits, teachers’ optimal instructional practices in textbook use could be identified in said studies. Such findings would potentially be useful in Teacher Education for pre- and in-service teachers to foster informed textbook-consumption practices and compensatory measures when needed for the students’ solid attainment of L2 proficiency.
The present multiple case study examines the adaptation techniques and the presence of FFI and MFI resulting from the implementation of such techniques and retained original textbook activities in Spanish English as a Foreign Language (EFL) classrooms, an under-researched context in textbook use. The identification of the adaptation techniques was performed with a coding scheme grounded in the data itself. FFI and MFI were measured with a scale specially devised for this study. The scale represents a potential tool to systematize such a measurement for both contexts of textbook content and use, and thus complements the scarce instruments found in the literature to date.
All in all, it is expected that, by broadening scholarly knowledge about textbook use patterns displayed by teachers, the findings of this study will contribute to raising much-needed awareness of the crucial implications of textbook use for students’ development of communicative competence in both fields of Teacher Education and ISLA.
2. Literature Review
2.1. Form-Focused and Meaning-Focused Instruction (FFI and MFI)
In accordance with the overarching goal of Communicative Language Teaching (CLT), the contemporary prevailing approach to Foreign Language Teaching (FLT), students should become competent users of the targeted language. In other words, they should be able to use it for communicative purposes, with sufficient intelligibility, fluency and accuracy as well as sociolinguistic and pragmatical ability. The practical implementation of CLT is not homogenous given the wide interpretations derived from its wide theoretical principles; however, there is general agreement on the classroom principles that should contribute to the development of the student’s communicative competence (Graves & Garton, 2017, pp. 447-448):
· language should be authentic and with an emphasis on meaning
· there should be a balance of fluency and accuracy
· both productive and receptive skills should be developed
· language forms are not excluded but should be presented in context
· materials should be authentic
· the target language should be used in the classroom by both teachers and learners whenever possible
· learning should be active and collaborative.
Therefore, in accordance with the aforementioned classroom principles, attention should be paid to two types of instruction: FFI and MFI. MFI (Focus on Meaning in Long’s [ (Long, 1991)] terms) is premised on the communicative nature of language. Traditional examples of MFI are the 1960’s French immersion programs in Canada, Krashen’s (Krashen, 1982) Natural Approach, the strong version of CLT and Task-based Language Teaching (TBLT). MFI supports acquiring languages in an implicit, natural way, without conscious awareness and intent (Krashen, 1982). There is no attempt to direct learners’ attention to any specific linguistic form for the purposes of acquiring and developing linguistic knowledge. In the FLT classroom, MFI can be provided in activities that elicit free, spontaneous, creative, use of the language, as well as receptive skills, especially of the extensive type (Criado, 2022).
The fact that the empirical research resulting from the French immersion programs in Canada revealed students’ high levels of fluency and confidence in using the language, but notoriously abundant formal errors (especially in grammatical, phonological and pragmatic features) provided indirect support for a form-oriented component in language pedagogy (Ranta & Lyster, 2018; Storch, 2018). In this respect, Long (Long, 1991, 1996) distinguished Focus on FormS and Focus on Form. Focus on FormS represented traditional classroom teaching, that is, comprising rule explanation and controlled practice of isolated forms extracted from a structural syllabus. Based on psycholinguistic principles (reassessed by Ellis, 2016, in terms of selective attention, cognitive comparison, timing of focus on form and working memory), Long (Long, 1991) rejected this approach in favor of Focus on Form, which he defined as “overtly draw[ing] students' attention to linguistic elements as they arise incidentally in lessons whose overriding focus is on meaning” (pp. 45-46). This approach to language pedagogy, exclusively reactive, was materialized in corrective feedback and improvised formal explanations. In a more flexible way, Spada (Spada, 1997) used the term FFI to refer to “any pedagogical effort which is used to draw the learners’ attention to language form either implicitly or explicitly” (p. 73). Ellis (Ellis, 2001) stated that FFI refers to “any planned or incidental instructional activity that is intended to induce language learners to pay attention to linguistic form” (pp. 1-2). Therefore, as Loewen and Sato (Loewen & Sato, 2017) asserted, FFI is a superordinate category opposed to MFI and it comprises both the subordinate categories of Focus-on-Form(s) and Focus-on-Form. These should be considered as “poles on a continuum, in which the ratio of attention to language form and meaning changes proportionally” (Loewen & Sato, 2017, p. 5). Also, FFI can be isolated or integrated (Spada & Lightbown, 2008; Valeo, 2018). In the former, students’ attention is directed to form before or after communicative-based activities, while in the latter such attention to form happens briefly as embedded in communicative-based activities.
Basically, the aim of FFI is to help students become aware of certain language forms that may go unnoticed due to their low saliency, such as definite articles or the third person singular in the present simple tense. However, as Ranta and Lyster (Ranta & Lyster, 2018) acutely claimed, FFI should not be confused with decontextualized grammar teaching to which students cannot perceive any close or useful relevance for their final goal of becoming competent L2 users. Activities in the communicative classroom should have face facility and be enjoyed by the students (McDonough & Sato, 2019). FFI can include both proactive (planned) and reactive as well as explicit and implicit options (Loewen, 2011). The more implicit alternatives (e.g., input enrichment, input enhancement and recasts) reflect a higher focus on meaning than on form, or, in other words, a lesser degree of attention to form. Given that in a formal context there will always be a certain focus on linguistic forms, even if subliminally, this characteristic makes such implicit alternatives especially suited to favor MFI in classrooms. More explicit options such as metalinguistic explanations naturally center on linguistic forms.
The benefits of FFI have been attested in several meta-analyses: Norris and Ortega’s (Norris and Ortega, 2000) seminal work, which included 49 studies published between 1980 and 1998; Goo et al. (Goo et al., 2015), which was based on 34 studies dated from 1999 and 2011, and Kang et al. (Kang et al., 2019), which drew on 54 studies published between 1980 and 2015. More recently, the meta-analysis conducted by Li and Sun (Li & Sun, 2023) with 28 quasi-experimental studies dated from 1991 to 2017 unveiled the higher benefits of explicit deductive instruction, operationalized as rule information and practice, on the one hand, and explicit inductive instruction (consisting of rule formulation and external guidance), on the other, over the three other types of explicit instruction distinguished (explicit deductive instruction with rule information but without practice, explicit inductive instruction involving rule formulation without external guidance, and explicit inductive instruction without both rule formulation and external guidance).
The consideration of both MFI and FFI is essential to ensure that students receive sufficient and optimal opportunities to develop the different types of L2 knowledge distinguished in Skill Acquisition Theory (e.g., Anderson & Schunn, 2000): declarative and procedural knowledge, usually coinciding with explicit (conscious) and implicit (subconscious) knowledge respectively (but see DeKeyser, 2009, for an excellent account in the SLA literature of the differences between declarative and explicit, procedural and implicit types of knowledge). Declarative explicit knowledge is factual knowledge; in L2 learning, it refers to the knowledge of linguistic forms (grammatical, lexical, phonological, orthographical) and of textual, pragmatic, socio-cultural features, or knowledge about language (DeKeyser, 2017). Procedural knowledge is factual instrumental knowledge (knowing how to do something). It becomes automatized (i.e., fast, effortless, unconscious and efficient; DeKeyser, 2017) in its ultimate point of development. In other words, proceduralized implicit knowledge is the type of knowledge that underlies the ability to communicate in the target language with accuracy and fluency in intercultural real-life situations. Nevertheless, as Loewen and Sato (Loewen & Sato, 2017, p. 4) claimed, “it is possible for learners to possess both types of knowledge of the same linguistic feature”. In fact, while learners are engaged in a communicative activity, they may become aware of certain linguistic features that they were not familiar with; consequently, they potentially develop both declarative and procedural knowledge within the same task. In correspondence with the strong interface stance, the cognitive sequence advocated in Skill Acquisition Theory promotes the attainment of declarative knowledge prior to procedural and automatized knowledge through repeated practice (DeKeyser, 2009). Accordingly, the application of Skill Acquisition Theory to L2 pedagogy roughly entails a sequence from accuracy to fluency. In terms of FFI and MFI, it is expected that the former will mostly develop declarative and preliminary or incomplete procedural knowledge, while the latter will mainly foster the attainment of full procedural knowledge and thus of automatized knowledge, resulting in fluent language use (which should be accurate too; Criado, 2022). To promote a solid development of L2 mastery, instruction should ensure the application of transfer-appropriate processing (TAP; Lightbown, 2008; Ranta & Lyster, 2018). This principle from Cognitive Psychology is premised on the fact that the conditions of language performance in learning should parallel those of the world outside the classroom as much as possible. In other words, supplying FFI alone will not suffice to train students for real-life communicative purposes.
A very useful pedagogical framework of FFI which aligns with Skill Acquisition Theory is that provided by Ranta and Lyster’s (Ranta & Lyster, 2018) classification of proactive FFI techniques: 1) input enhancement, which is considered a less implicit technique than input enrichment and it is aimed at facilitating the noticing of certain flooded items (Schmidt, 1990), by highlighting them in the oral or written input; 2) metalinguistic explanations, targeted at developing awareness and, therefore, explicit knowledge of the linguistic features, either deductively or inductively, as a prior step for the optimal development of implicit knowledge; and (iii) practice. The latter ranges from more controlled (but ideally meaningful) types targeted at strengthening accuracy to freer types such as communicative practice (e.g., communicative drills), which entails using already pre-taught language forms in conveying and processing new or unpredictable information. As Criado (Criado, 2022) explained, in the FLT classroom, the attainment of automatization can be possibly attained in two ways related to the application of TAP: with focused tasks—the freest practice option in FFI, which is designed to elicit the understanding and use of certain predetermined linguistic features in a way as similar as possible to real-life communication processes–and MFI or purely communicative activities, whether receptive such as extensive reading or listening, or productive (for instance, a free discussion or essay writing).
In short, as Storch (Storch, 2018, p. 5) summarized, specialized literature advocates “a communicative approach to L2 instruction which integrates a focus on meaning and on form”. Consequently, it is of utmost importance, not only to study the degree to which textbooks include FFI- and MFI-based activities to potentially trigger the suitable development of types of L2 knowledge, but also to examine the extent to which teachers manipulate the implementation of FFI and MFI existing in the prescribed textbooks when using them in their classrooms.
2.2. Textbook Use and Adaptation
Textbooks are also called coursebooks (Zhang et al., 2022) and both terms will be indistinguishably used in the present study. Published textbooks are the traditional and most frequent types of material used by teachers in their ordinary classroom practice and lesson planning. Of course, teachers may also draw on resources which are assigned a pedagogic role, such as a YouTube video (see Mishan & Timmis, 2015, for the distinction between resources and materials, and Luque-Agulló, 2022, for an exhaustive list of materials and resources that teachers may choose from). Traditional conceptualizations of materials have lately been complemented with out-of-class learning opportunities provided, for instance, by technology (Harwood, 2021).
My focus will be on commercial textbooks, both global or international coursebooks—that is, from major English as a Language Teaching/ELT publishers—and those nationally produced (Mishan, 2022). In the field of materials research (Gray, 2012), this study is framed within “textbook consumption” (Harwood, 2014) or “materials use” (Graves, 2019), to which limited scholarly attention has been paid, as opposed to textbook content analysis (Graves, 2019; Li & Xu, 2020; Rathert & Cabaroğlu, 2021; Schwab, 2022, etc.). In this article, the terms “textbook use” and “textbook consumption” will be used as synonyms. Harwood (Harwood, 2022) also distinguished two main strands of research within textbook consumption: learning gains as derived from the textbooks being consumed (on which very little research has been carried out; see Harwood, 2022, for a review of the two studies he mentioned— Boers et al., 2014, 2017), and teachers’ patterns of use, involving the analyses of their adaptations and reliance on the materials. This study will focus on the latter strand, corresponding to Graves’ perspective of use by instructors: “how learners and teachers actually use materials (as distinct from how materials should or could be used)” (p. 78). Certainly, teachers are crucial agents who mediate how the textbook content is conveyed, whether unchanged or adapted, in order to “minimize mismatches between coursebook content and educational context requirements” (Zhang et al., 2022, p. 2).
Harwood (Harwood, 2021) emphasized the flexible nature of the approach to materials use, “showing how skillful teachers make many pre- and in-lesson decisions as to how to tailor the materials to best fit the unfolding interaction between learners and the materials minute by minute and activity by activity in the classroom” (p. 180). A myriad of variables affecting such behavioral patterns and their impact has been uncovered in the empirical, mostly qualitative literature available: the teachers’ sense of self-confidence, pedagogical knowledge, experience and motivation; their students, logistic constraints, institutional requirements in the curriculum, or the washback effect, etc. (Zhang et al., 2022).
Table 1 categorizes the adaptation techniques distinguished in the most recent specialized literature and which other authors have recurrently resorted to (e.g., Marcos Miguel, 2015; Rathert & Cabaroğlu, 2021; Schwab, 2022, etc.). Of course, the teacher can opt for not adapting the activities, which corresponds to selection in McGrath’s (McGrath, 2013, 2016) terms. For a diachronic overview of the terminology underlying adaptation techniques, see McGrath (McGrath, 2013, pp. 64-65). The account following Table 1 will be based on the classifications provided by McDonough et al. (McDonough et al., 2013, chapter 4) as well as by McGrath (McGrath, 2013, chapter 6, and 2016, chapters 4 and 5).
McDonough et al. (McDonough et al., 2013) outlined adding (including extending and expanding), deleting (covering subtracting and abridging), modifying (comprising rewriting and restructuring), simplifying and reordering. McGrath (McGrath, 2016) referred to selection, deletion, addition and change as evaluative processes of the coursebook for lesson planning purposes and he only considered addition and change as representing adaptation.
![]()
Table 1. Adaptation techniques distinguished in the specialized literature.
For McDonough et al. (McDonough et al., 2013), extension implies a quantitative addition of the same content (e.g., more items to practice in a pronunciation, grammar or vocabulary activity), while expanding involves a qualitative change in the methodology too. These authors illustrated expanding with the addition of work on sentence stress and rhythm to the pronunciation practice of the textbook, focused on individual sounds and minimal pairs; providing a listening practice activity that parallels a previous reading text in linguistic and content input, and adding a discussion section at the end of a unit to help consolidation of the previously introduced new grammar forms in a contextualized way. McDonough et al.’s examples of expanding seem to evoke McGrath’s (McGrath, 2013, 2016) supplementation, which he did not regard as an adaptation technique per se. Supplementation refers to the provision of an additional exercise “whether copied from another source or devised by the teacher” (McGrath, 2013, p. 72). Besides extension, McGrath (McGrath, 2013, 2016) also highlighted two further addition techniques: extemporization and exploitation. The former consists of “examples, explanations, paraphrases offered spontaneously in response to a predicted or perceived learner problem” (McGrath, 2013, p. 65). Thus, when this technique is applied to language forms, it could be considered a preemptive approach to FFI (Ellis, 2001). The latter refers to creative uses of material already existing in the textbook for purposes not intended by the original authors. Among other examples, McGrath (McGrath, 2013) mentioned using a picture that accompanies a text as a stimulus to activate vocabulary or predict content, or asking students to use the questions of the text as models so that they devise their own questions.
As for deleting, McDonough et al.’s (McDonough et al., 2013) technique of subtracting implies a mere quantitative reduction of the material. These authors illustrated this technique with the decrease of the number of items to be practiced in a minimal-pairs pronunciation exercise when it is not necessary to cover all of them (perhaps because the students share their first language or L1, which facilitates the identification and production of certain L2 sounds). Another example provided by McDonough et al. concerns the reduction of content in the full textbook, such as some language functions, on account of students’ needs. Contrary to subtracting, abridging entails a qualitative change as well. McDonough et al. exemplified it with the deletion of the discussion section at the end of each unit given the limited students’ level of proficiency, and the elimination of long grammatical explanations accompanying each functional unit in a communicative course whose students’ reason for enrolment is instrumental in nature.
Modification is related to “a ‘modality change’, to a change in the nature or focus of an exercise, or text or classroom activity” (McDonough et al., 2013, p. 74). These authors considered restructuring as changes in the structuring of a class, that is, classroom management. The examples provided by McDonough et al. are the reassignment of specific roles of a roleplay activity to different students simultaneously in a large class that shares the same L1, and offering students the possibility of applying topics of their interest when practicing a certain structure, so that the activity becomes more authentic. Rewriting refers to modification of the linguistic input. McDonough et al. illustrated it with the teachers’ conversion of a reading text into a listening text delivered orally by themselves, in an attempt to compensate the shortage of listening material; changing the English names of a story with local ones which are felt to be more appropriate for a specific classroom from a cultural perspective, or turning the literal comprehension questions in a reading activity to interpretation questions, etc.
Although McDonough et al. (McDonough et al., 2013) separated simplifying from rewriting due to the recurrent attention it has received in literature, they acknowledged that it is a rewriting activity. As these authors indicated, the alternatives of simplifying are varied: reducing sentence length, or transforming structures (e.g., from passive to active; from reported to direct speech), or controlling the ratio of new lexical items to those already learned. In other words, the purpose of simplification is to make materials easier for learners (McGrath, 2013), by modifying the input instead of undertaking procedural changes in the activity. However, authors have raised concerns about the potential undesirable effects of simplification on the amount of linguistic input and naturalness of texts students should be exposed to, as well as their difficulties in comprehension due to the elimination of many redundant features (e.g., Le, 2011). In the same section as simplification, McGrath considered complexification—augmentation of the level of difficulty—and differentiation, which attempts to cater for learner differences (e.g., proficiency level or learning styles).
Regarding replacement, McGrath (McGrath, 2016) acknowledged that it “will never fulfil exactly the same purposes as the original” (p. 68), but it can compensate teachers’ conscious omission of some material perceived to be unsuitable for learners on account of their age, interests, cultural background and background knowledge. Examples provided by McGrath were the replacement of the map of a town unknown to the students with another one of their own town, changing the character of a famous British person to another one the students are familiar with, etc.
Concerning reordering, this can entail changing the order of items or content within a unit, or even “taking units in a different sequence from that originally intended” (McDonough et al., 2013, p. 75). The examples offered by these authors seem to refer to the second context mostly. For example, joining all the language structures that deal with the notion of future (will, going to, simple present and the present continuous), or rearranging excessive new grammar points for a particular language function based on learners’ proficiency level.
The aforementioned terminology, though always clear and sensible in theory, sometimes is not so unequivocal when one examines the examples proposed to illustrate each adaptation technique or analyzes activities implemented by teachers in their classes. This fact is not meant to be a negative criticism of the frameworks provided by McDonough et al. (McDonough et al., 2013) and McGrath (McGrath, 2013, 2016), but an invitation to reflect on the complexity of the issue. For instance, the parallelism between McDonough et al.’s expanding and McGrath’s supplementation has already been noted. Restructuring may imply not only a change in classroom dynamics (McDonough et al.), but also a change in the modality of the input, such as when a listening text is converted into a reading one due to the absence of audio equipment or a washback effect (Rathert & Cabaroğlu, 2021). In the case of reordering, besides content (which seems to be the major focus of McDonough et al.’s examples), its very name also allows us to consider the possibility that this adaptation technique can affect activities explicitly, both within and across them. For instance, regarding the first case, Rathert and Cabaroğlu (Rathert & Cabaroğlu, 2021) offered the example of changing the correct order of the paragraphs in a text so that learners order them correctly. Reordering can also refer to resequencing entire activities within a unit (see the section called Analysis of the Adaptation Techniques in Textbook Use). Finally, McGrath’s (McGrath, 2013, 2016) category of exploitation within addition could arguably be assimilated to a (sophisticated) type of change, or McDonough et al.’s restructuring. The exploited activity suffers a radical modification in its procedure of implementation. Certainly, determining a threshold of creativity to decide whether the adaption technique refers to modification or exploitation appears to be a complex task.
Besides the definition of the adaptation techniques that teachers may apply in the classroom, teachers’ roles regarding textbook use as derived from the quantity and nature of their adaptations have been considered too (Shawer, 2010, 2017). On the basis of his own classroom observations of 10 EFL college teachers using textbooks, Shawer (Shawer, 2010) identified three types of approaches to curriculum use: “curriculum fidelity (curriculum-transmission); adaptation (curriculum-development) and enactment (curriculum-making)” (p. 174). Behavioral patterns revealed by curriculum transmitters include adherence to both the textbook’s content and sequencing of elements, as well as using a single textbook in a lesson-by-lesson and page-by-page way—thus, in a predictable way. Curriculum developers use certain strategies such as a flexible use of the textbook by adapting lessons or tasks, supplementing lessons or skipping tasks. Finally, curriculum makers do not use the textbook but multiple sources of input instead, based on the assessment of their students’ needs, with a large selection of topics not corresponding to the officially prescribed materials. Overall, curriculum developers and makers consider the textbook a resource, while curriculum transmitters treat it as a script (Tomlinson, 2022). Following Schwab (Schwab, 2022), in this study the term “curriculum” will be replaced with “textbook” or “coursebook”, as in FLT “the term curriculum is often used to refer to the mission of the programme and not for the materials used in class” (p. 106).
2.3. Empirical Studies on Classroom Textbook Use
This section deals with studies which examined classroom textbook use and relied on classroom observation as one instrument of data collection. Thus, it will not cover those studies which reported self-reported data extracted from questionnaires (Bolster, 2015; Zhang et al., 2022) and interviews (Bosompem, 2014).
2.3.1. Methodological Considerations
The educational contexts and countries subject to study have been varied. Within Primary Education, Loh and Renandya (Loh & Renandya, 2015) drew on Singaporean schools, while Swiss centers were studied by Schwab (Schwab, 2022). Aftab (Aftab, 2022) focused on Pakistani middle schools (both public and private). As for Secondary Education, schools from Egypt and Albania were respectively examined by Abdel Latif (Abdel Latif, 2017), Seferaj (Seferaj, 2014); Vietnam classrooms were the focus of Nguyen et al. (Nguyen et al., 2018) and Tran et al. (Tran et al., 2023). Tertiary education was covered in the United Kingdom (Grammatosi & Harwood, 2014); Saudi Arabia (Menkabu & Harwood, 2014); the United States (Guerrettaz & Johnston, 2013; Marcos Miguel, 2015); Oman (Tasseron, 2017); Vietnam (Dao & Newton, 2021); China (Li & Xu, 2020) and Turkey (Rathert & Cabaroğlu, 2021). Shawer (Shawer, 2017) examined private language colleges in the United Kingdom. To the best of my knowledge, there is no research on non-formal education. The teachers’ experience ranged from two months (Nguyen et al., 2018) to more than 20 years (Li & Xu, 2020; Nguyen et al., 2018; Seferaj, 2014; Shawer, 2017). All the previous studies dealt with EFL except for Marcos Miguel (Marcos Miguel, 2015), who studied Teaching Assistants of Spanish. From the information included in the studies, native teachers participated in only two of them (Grammatosi & Harwood, 2014; Shawer, 2017). The least frequent students’ levels of the observed classes in those studies which explicitly reported this information were upper-intermediate and advanced (Shawer, 2017). The remaining ones ranged from elementary (Dao & Newton, 2021; Grammatosi & Harwood, 2014; Menkabu & Harwood, 2014; Tasseron, 2017) to pre-intermediate (Menkabu & Harwood, 2014; Rathert & Cabaroğlu, 2021; Shawer, 2017; Tasseron, 2017) and intermediate levels (Marcos Miguel, 2015; Rathert & Cabaroğlu, 2021; Shawer, 2017).
A broad variability of some crucial methodological aspects regarding adaptation techniques can be observed in the previous studies:
1) Authors did not homogeneously code the classroom data: addition, reordering, modification and simplification (Marcos Miguel, 2015); adaptation, creation and retention (Nguyen et al., 2018); addition, omission and retention (Dao & Newton, 2021); addition (extension, exploitation and extemporization), reordering, modification, omission and retention (Schwab, 2022); addition (extemporization, expanding, exploitation and supplementation), modification (restructuring, rewriting, reordering), omission (abridging and subtracting) and substitution (Rathert & Cabaroğlu, 2021). Loh and Renandya (Loh & Renandya 2015) is the only study that followed Tomlinson and Masuhara’s (Tomlinson & Masuhara 2004) framework (plus, minus and zero categories, which, as Loh & Renandya stated: plus refers to an increase in quantity, minus involves a decrease in quantity and finally zero, referring to modification without any change in quantity).
2) Although the general tendency in terms of the object of study was the textbook, Schwab (Schwab, 2022) indicated that she used the term “textbook” as comprising both the individual coursebook and the whole package; in fact, she reported the number of adapted activities out of the total number of the activities from the resources that teachers resorted to (the pupil’s book, the activity book, the worksheets and the e-Book). Loh and Renandya’s (Loh & Renandya, 2015) “artifact analysis” included “lesson plans, teaching resources, school-based curriculum plans, school worksheets, teacher-prepared worksheets” (p. 105).
3) Most studies where more than one teacher was observed examined the use of the same textbook by different instructors (e.g., Dao & Newton, 2021; Li & Xu, 2020; Loh & Renandya, 2015; Marcos Miguel, 2015; Menkabu & Harwood, 2014; Rathert & Cabaroğlu, 2021; Schwab, 2022; Tasseron, 2017; Tran et al., 2023), as opposed to Shawer (Shawer, 2017). Different levels of the same series were studied too (e.g., Abdel Latif, 2017; Aftab, 2022, and Nguyen et al., 2018). Though the full content of the textbooks was usually considered for examination in classroom use, some studies dealt with specific parts or types of activities. For instance, Abdel Latif (Abdel Latif, 2017) analyzed the teachers’ use of the grammar sections, while Nguyen et al. (Nguyen et al., 2018) focused on oral tasks.
4) Concerning the nature of the data analyzed for the classroom adaptations, many studies reported qualitative information (e.g., Aftab, 2022; Grammatosi & Harwood, 2014; Guerrettaz & Johnston, 2013; Li & Xu, 2020; Menkabu & Harwood, 2014; Seferaj, 2014; Shawer, 2017; Tasseron, 2017). Some others offered quantitative data (e.g., Abdel Latif, 2017; Dao & Newton, 2021; Loh & Renandya, 2015; Marcos Miguel, 2015; Nguyen et al., 2018; Rathert & Cabaroğlu, 2021; Schwab, 2022). Tran et al. (Tran et al., 2023) included quantitative information about the adaptation techniques found in the lesson plans the teachers devised prior to their classes.
5) The way some studies reported the quantitative results concerning the adaptations performed in class does not seem to allow for a straightforward comparison in this respect. For instance, while most authors included the number of the textbooks’ original activities and that of the teachers’ adapted ones (Dao & Newton, 2021; Marcos Miguel, 2015; Nguyen et al., 2018; Schwab, 2022), Loh and Renandya (Loh & Renandya, 2015) reported the percentages of the presence and absence of adaptation techniques, whose addition equaled the number of lessons observed. Rathert and Cabaroğlu (Rathert & Cabaroğlu, 2021) offered the frequencies of adaptations of two teachers without including the total number of activities performed in their lessons. Thus, the number of retained activities (whether any at all) is unknown. Moreover, the frequencies of the added activities in Dao and Newton (Dao & Newton, 2021), Marcos Miguel (Marcos Miguel, 2015) and Schwab (Schwab, 2022) cannot be rendered into percentages, since the total number of observed activities was not included to compare it against the total number of the textbooks’ activities to which the teachers applied the adaptation techniques. Therefore, in these cases, only the frequencies of the activities added by each teacher can be considered.
6) Most of the studies that reported quantitative data of the adaptation techniques did not specify whether such types were coded as mutually exclusive among them or not, which is another factor that impedes a direct comparison of the results between studies. Only Rathert and Cabaroğlu (Rathert & Cabaroğlu, 2021) explicitly stated that “All coding categories were operationally defined to ensure mutual exclusivity” (p. 9). Conversely, the data in the current study showed that different adaptation techniques could affect the same activity (see the section called Analysis of the Adaptation Techniques in Textbook Use). As it may also be inferred in the case in Marcos Miguel (Marcos Miguel, 2015) and Schwab (Schwab, 2022), one activity could be both modified and reordered. Nevertheless, this conclusion remains speculative since these two authors and those of the other studies did not specify the exclusivity or inclusivity of their coding framework.
7) Finally, a critical methodological weakness in most studies is that they did not report reliability measures for the coding of the adaptation techniques. Harwood (Harwood, 2017, 2022) already advised on the importance of implementing this type of analysis in textbook consumption studies. Nguyen et al. (Nguyen et al., 2018) and Dao and Newton (Dao & Newton, 2021) conducted an inter-rater reliability analysis for the coding of the features of task design for the former and degree of communicativeness and task-likeness of the activities for the latter. Also, and just as important, no study that included quantitative data of the adaptation techniques reported any inter-rater reliability measures for the segmentation of the activities observed. Both points will be addressed in the Analysis of the Adaptation Techniques in Textbook Use section.
2.3.2. Patterns in Classroom Textbook Use
Given the methodological caveats indicated before, the following account will consider the quantitative patterns of adaptation found in the previous studies from a global perspective. Also, no specification will be made about whether the results stem from the adaptations of the textbooks’ original activities or whether they refer to the activities set in the classroom by the teachers.
The analysis of the results obtained in the different studies revealed that, from a quantitative perspective, adaptation was frequently undertaken by all the teachers observed in different degrees (Dao & Newton, 2021; Loh & Renandya, 2015; Marcos Miguel, 2015; Nguyen et al., 2018; Schwab, 2022). Shawer (Shawer, 2017) reported that the amount of textbook adaptation carried out by each pair of teachers clearly corresponded to their assigned profile and style, with the highest one corresponding to the textbook makers. With a detailed qualitative analysis, in which he used the constant comparative method to examine the data from the interviews and participant observation, Shawer identified the teachers’ teaching styles as systematic, occasional improvisers and permanent improvisers (respectively corresponding to textbook transmitters, developers and makers).
Other qualitative studies have also revealed that teachers resorted to adaptation in a more or less recurrent way (e.g., Grammatosi & Harwood, 2014; Li & Xu, 2020; Menkabu & Harwood, 2014; Seferaj, 2014; Shawer, 2017; Tasseron, 2017). However, Aftab (Aftab, 2022) indicated that the 12 EFL teachers observed hardly adapted the textbooks.
Omission of activities was the most frequent adaptation technique among the studies which reported frequencies (Dao & Newton, 2021; Marcos Miguel, 2015, Schwab, 2022). Abdel Latif (Abdel Latif, 2017) also reported that out of the 168-192 stipulated grammar activities to be taught in the classes observed, nearly half of them were omitted. The teachers in Loh and Renandya (Loh & Renandya 2015) followed an opposite trend, since the most common type of adaptation was the “addition” category. In the case of Nguyen et al. (Nguyen et al., 2018), no data were computed for omission, but for adaptation and creation, the latter being the most common adaptation technique. A persistent trend in specialized literature is the scarce reference to the addition techniques of extension, expansion, extemporization and exploitation, whose presence was low, especially exploitation (Rathert & Cabaroğlu, 2021; Schwab, 2022). However, three out of the four teachers observed in Schwab (Schwab, 2022) used extemporization very frequently, shaped as the translation of instructions into the learners’ L1, paraphrasing and/or explanation of instructions, and the teacher’s resorting to one or two students for the demonstration of an activity prior to its completion. Modification was identified by Marcos Miguel (Marcos Miguel, 2015) and Schwab (Schwab, 2022), which was the second most frequent adaptation technique in both studies. Different types such as replacement or restructuring, complexification or differentiation were not distinguished. The same authors detected small amounts of reordering, which they did not include within modification.
The generally most frequent presence of omission of activities in comparison with other adaptation techniques (despite Loh & Renandya’s [ Loh & Renandya 2015] and Nguyen et al.’s [ Nguyen et al., 2018] exceptions) does not seem to align with the general patterns reflected in the qualitative studies. For instance, the teacher in Grammatosi and Harwood (Grammatosi & Harwood, 2014) largely replaced and supplemented the official textbook prescribed, which ascribed him to the category of textbook developer. The teacher in Guerrettaz and Johnston (Guerrettaz & Johnston, 2013) also supplemented the grammar textbook, despite his implicit feeling that the legitimate curriculum was represented by the coursebook. Menkabu and Harwood (Menkabu & Harwood, 2014) reported that addition was the most common teaching technique in order to maintain their students’ interest. Furthermore, the teachers’ claim in the interviews that they frequently modified the textbook was not corroborated in the classroom observations. Seferaj (Seferaj, 2014) frequently supplemented the textbook with grammatical materials but not with extra vocabulary activities.
Besides addition, replacement of perceived culturally inappropriate cultural content was the most frequent adaptation technique in Tasseron (Tasseron, 2017). The three teachers studied in Li and Xu (Li & Xu, 2020) revealed a diverse adaptation behavior, which comprised supplementation, modification of instructions or changing the intended pedagogical goals of the original materials; nevertheless, the teachers’ overreliance on the textbook reflected an authoritative role assigned to it.
As stated in the section called Textbook Use and Adaptation, the reasons that teachers alleged to explain or justify their perspective of textbook use in general or about specific episodes (with stimulated recalls as in Dao & Newton, 2021; Nguyen et al., 2018; Seferaj, 2014 and Rathert & Cabaroğlu, 2021) were varied. Such motives included learners’ proficiency and motivation, time constraints, the washback effect, the teachers’ own teaching preferences, etc. For example, time limitations accounted for the frequent omission of activities in Marcos Miguel’s (Marcos Miguel, 2015) teaching assistants. Abdel Latif (Abdel Latif, 2017) found that teachers frequently omitted and replaced the inductive and oral communicative activities in the prescribed textbooks because such activities did not comply with their own teaching styles, as they preferred a deductive approach which would foster their active role. Conversely, the teacher in Grammatosi and Harwood (Grammatosi & Harwood, 2014) largely supplemented the prescribed textbook especially due to his perception that the level of the language was difficult for his students and that the topics were not relevant for them, as well as logistic factors (such as new students’ enrolment in the middle of the course). Seferaj’s (Seferaj, 2014) teacher consciously adapted the textbook on account of her students’ needs, given her long-time experience and knowledge of her context. This entailed a partial departure from the communicative approach followed in the textbook; for instance, replacing fluency interactive activities with controlled teacher-student exchanges due to her students’ different levels and resulting difficulties in establishing suitable classroom dynamics. As opposed to this pattern, the teachers observed in Ngyuen et al. (Ngyuen et al., 2018) frequently adapted the textbooks’ activities to render them more authentic to their students. Similarly, the instructors in Li and Xu (Li & Xu, 2020) drew on a varied range of adaptation techniques in an attempt to comply with their students’ likes, needs and proficiency. As can be seen, in general, one main factor rooted in teachers’ classroom adaptations is compliance with the local context and an attempt to suit students’ needs, regardless of their level and educational stage. In other words, the immediate context is the crucial mediating variable for how teachers use textbooks since, as Li and Xu (Li & Xu, 2020) claimed, “teachers’ enactment of materials is interpretive, dynamic, and interactive” (p. 9). However, not all teachers appeared to resort to adaptation in differing degrees. Some of them largely complied with the textbook’s content and sequence of activities, due to time and exam pressures (e.g., Aftab, 2022; Shawer’s [ Shawer, 2017] textbook-transmitter teachers).
Although it is complicated to establish a general pattern of textbook use as reflected in specialized literature, it seems that the least frequent profile is that of textbook maker, while the most prevalent one appears to be that of a textbook transmitter—with both rigid and moderate degrees, the latter mostly resulting from a relatively high presence of activity omission. It is relevant to point out that the teachers’ experience, however, cannot be considered a defining variable for the emergence of the textbook-user profile less inclined to transmission. This points to the importance of teachers’ training and professional development as well as teaching styles (Shawer, 2017). Such circumstances seem to be especially relevant for the fourth profile distinguished by Schwab (Schwab, 2022): “textbook deviator” (with little textbook awareness and ad-hoc decision-making behavior).
Regarding the focus of this study on FFI and MFI, as far as I can ascertain, only three studies have examined textbook use from this perspective. Their degree of exhaustiveness is notably varied: from the detailed analysis in Dao and Newton (Dao & Newton, 2021) to a moderate reference in Seferaj (Seferaj, 2014) and a brief allusion in Tran et al. (Tran et al., 2023). Dao and Newton (Dao & Newton, 2021) drew upon Littlewood’s (Littlewood, 2004, p. 322) categorization of activities following “the continuum from focus on forms to focus on meaning”. This was renamed as the “continuum of communicativeness of activities” by Dao and Newton (p. 105), with which they analyzed activities from the textbook and those implemented by the three teachers whose classes were observed. Littlewood (Littlewood, 2004) distinguished five categories to which activities can be ascribed in the following ascending order of communicativeness: non-communicative learning, pre-communicative language practice, communicative language practice, structured communication and authentic communication (see the Analysis of the Measurement of FFI and MFI section). Dao and Newton (Dao & Newton, 2021) assigned a low communicative value to the activities they classified within the first two categories and a high communicative value to those present in the last three categories. Considering the idiosyncratic attention to language forms, in different degrees, of classroom teaching (see the Form-focused and Meaning-focused Instruction [FFI and MFI] section), low-communicative value activities could be assimilated to FFI, while those with a high communicative value arguably represent a range from FFI (communicative practice and structured communication) to MFI (authentic communication). The analysis of the two sets of activities revealed the prevalence of the low communicative value. Seferaj (Seferaj, 2014) studied the implementation of a communicative textbook by an experienced Albanian EFL teacher in Albanian secondary schools. Similar amounts of time were devoted to form-focused and meaning-focused activities. Although the observed teacher tried to integrate communicative principles in her lessons (using real-world tasks and activation of background knowledge before carrying out receptive activities), she also performed several adaptations that fostered teacher-led instruction and more emphasis on the controlled manipulation of structures. Finally, Tran et al. (Tran et al., 2023) reported that about one fifth of the original meaning-focused textbook activities were transformed into form-focused ones by the teachers when using the textbooks in their classes. These three studies highlighted the important decrease of the communicative orientation of the lessons observed due to the nature of the teachers’ adaptation and the already high number of form-focused activities in the textbooks. From the fragments of the teacher’s interview quoted in Seferaj (Seferaj, 2014), it can be inferred that she relied on the traditional PPP model of activity sequencing, which Dao and Newton (Dao & Newton, 2021) and Tran et al. (Tran et al., 2023) explicitly referred to as the approach revealed in the classroom observations (cf. Criado’s [ Criado, 2022] account of the “contemporary version of PPP”).
3. Rationale and Research Questions
Three major conclusions can be ascertained from the previous literature review: the last three methodological areas of improvement described in the Methodological Considerations section (referring to heterogeneous ways of selecting and coding the data, which rendered the comparison of results difficult, plus the absence of inter-rater reliability measures); the intricate pattern of findings regarding textbook use in L2 classes found, with the profile of textbook maker being the less frequent one; the scarce reference to the measurement of FFI and MFI in L2 classrooms as a result of textbook use and derived adaptations, as well as the need to develop systematized procedures to measure both types of instruction. It is relevant to determine how teachers deviate from textbook content patterns and the effects of their textbook use patterns on the suitability of their instructional practices to foster communicative competence in both an accurate and a fluent way.
This multiple case study attempts to contribute to ongoing research on textbook use by addressing both the aforementioned gap regarding the measurement of FFI and MFI and methodological limitations. Accordingly, it provides a comprehensive account of the patterns of classroom textbook use, both regarding the teachers’ adaptation techniques and the values of FFI and MFI as resulting from the application of said techniques. Such values were obtained with a scale purposefully devised for this research. Furthermore, the present multiple case study took place in three Spanish EFL classrooms—which, to the best of my knowledge, represent a context where this topic (textbook use and resulting FFI and MFI) has not been previously researched.
The following three research questions guided this study:
Research question 1: To what extent are there any differences in the percentage of adapted activities when instructors are teaching three different classes of the same linguistic level?
Research question 2: To what extent are there any differences in the percentage of adaptation techniques when instructors are teaching three different classes of the same linguistic level?
Research question 3: To what extent are there any differences in FFI and MFI between the textbooks’ original activities and those implemented by the instructors while teaching three different classes of the same linguistic level?
4. Method
4.1. Research Approach
The research approach adopted in this work was “multiple case study” (Stake, 2005) since, as Stake himself stated, “illustration of how a phenomenon occurs in the circumstances of several exemplars can provide valued and trustworthy knowledge” (pp. 458-459).
4.2. Context and Participants
I contacted several educational centers in southern Spain to request participation and explained the research purpose and conditions to the potential participating teachers of the centers whose Heads agreed to hold an in-person meeting with me. The purpose of these meetings was to provide more information about the outlined aspects (see the Classroom Observation section). After such meetings, only five teachers volunteered to participate. Since two teachers taught different levels from the other three, the final sample comprised three teachers and their corresponding B1-level classes (Council of Europe, 2001). Each teacher belonged to a different center, which will be labeled as Center A, Center B and Center C. The teachers will be referred to as Teacher 1, Teacher 2 and Teacher 3. Teacher 1 taught at an Official School of Languages, while the two other teachers taught students enrolled in their first year of Upper (non-compulsory)-secondary education.
Official Schools of Languages in Spain, framed within a non-university special regime, are publicly funded language schools. Students from the age of 16 and onwards can study modern languages and Spanish as a Foreign Language; besides, these centers also admit 14-year-old secondary-school adolescents enrolled in a first foreign language course at school whose targeted language is different from that chosen at the Official Schools of Languages. At the time of the data collection, the Official Schools of Languages were ruled by the Royal Decree 1629/2006 at a national level (Spanish Ministry of Education and Science, 2007) and by the specific regional decree corresponding to each one of the 17 autonomous communities, Ceuta and Melilla. The Royal Decree 1629/2006 established curricula descriptions given by the Council of Europe (Council of Europe, 2001) and which were aimed at level A2 (basic level), B1 (intermediate level) and B2 (advanced level), with either a 5-year or a 6-year period depending on the region. As a complement, C1 and C2 courses could be studied. Nowadays, the Official Schools of Languages also run C1 and C2 courses upon completion of which official certificates are issued.
The curriculum of Compulsory- and Upper (non-compulsory)-secondary education was ruled by the Royal Decree 1105/2014 at a national level (Spanish Ministry of Education, Culture and Sport, 2015) and, similarly to that of the Official Schools of Languages, the regional educational authorities of each autonomous community, Ceuta and Melilla were allowed to adapt certain specific aspects to complement the syllabus of compulsory subjects and to establish that of the specific ones, add to the evaluation criteria of the subjects, specify the number of contact hours of each one of them as well adapting the teaching methods. The Royal Decree 1105/2014 (as well as the derived regional legislation) incorporated the action-oriented approach and types of language activities as described by the Council of Europe (Council of Europe, 2001), that is, based on understanding and production (expression and interaction) of oral and written texts (Spanish Ministry of Education, Culture and Sport, 2015, pp. 422-423). Although no specific levels were mentioned as the target of each year (contrary to the regulation of the Official Schools of Languages), the wording of the descriptors of the evaluation criteria of the first and second-year of Upper-secondary education approximately seemed to converge to a B1 level. At the time of data collection, the students of the Official Schools of Languages received 4.5 weekly hours distributed over two days, while the students of Upper-secondary education attended four EFL sessions per week totaling four hours.
The teachers were all female Spanish native speakers. They were 35, 37 and 40 years old and they had 7, 8 and 10 years of experience. In the class of the first teacher, there were 12 male and 13 female students whose mean age was 35.7 (SD = 6.15). Ten males and 19 females (mean age = 16; SD = 0.5) attended the session taught by Teacher 2. Finally, Teacher 3 taught nine males and 15 female students (mean age = 16.7; SD = 0.8). All the students were Spanish native speakers.
4.3. Textbooks
After each session observed, I talked to the teachers and they confirmed that they had largely resorted to the textbooks for their teaching material in each session. Thus, for the purposes of analysis, the only material considered was the students’ coursebooks. Also, it should be taken into account that the instructions from the corresponding teachers’ guides or the teachers’ lesson plans were not examined.
In order to avoid potential conflicts arising from the analysis of the textbooks’ content, both their titles and publishers’ names will be omitted and the information referring to them will be as restricted as possible. The coursebooks selected will be referred to as Textbook I, Textbook II and Textbook III, respectively corresponding to teachers 1, 2 and 3.
Textbook I was the intermediate student’s coursebook of a series published by a global international brand. The back cover included its targeted level with a graph: comprehensive coverage of B1 and commencement of B2 (Council of Europe, 2001). Textbook I corresponded to the weak version of CLT as opposed to the strong one, following the distinction established by Howatt (Howatt, 1984): “the former could be described as ‘learning to use’ English, the latter entails ‘using English to learn it’” (p. 279). Accordingly, Textbook I offered a multi-component syllabus, comprising sections on the four skills and on the language systems of vocabulary, grammar and pronunciation, as well as pages on “real-life” English (covering functions and vocabulary practice). The textbooks’ authors claimed to offer balanced work on all such areas; besides, the final part of the coursebook also offered more related information and practice. Culture was intermingled in the units and the textbook included items related to both types of culture distinguished by Paige et al. (Paige et al., 2003): “Big C cultures” (e.g., literary pieces and history fragments) and, to a larger degree, “Little c cultures” (contemporary lifestyle and values, etc.).
Textbooks II and III were two different titles released by the same national publisher. The publisher, following its specialization in the Spanish market, stated on the back cover of Textbook II that it covered all the official requirements for Bachillerato (Spanish term for Upper-secondary Education). The textbook used by Teacher 3 explicitly included a reference to the level targeted, also by means of a graph: an almost complete B1 level. For the purposes of this study, the slightly higher level of Textbook I in comparison with that of Textbooks II and III was not considered to be relevant. The back covers of Textbooks II and II contained information about the availability of certain teachers’ resources such as material concerning the EFL section in the Spanish University Entrance Exam. In a similar way to Textbook I, both coursebooks complied with the weak version of CLT (Howatt, 1984). Therefore, they also offered a multi-component syllabus, comprising sections on the four skills and the language systems of vocabulary and grammar. The speaking sections practiced functions such as giving advice or comparing pictures and they also included explicit work on pronunciation. As opposed to Textbook I, culture was explicitly included in the table of contents as a separate section, with items that reflected Paige et al.’s (Paige et al., 2003) types of culture in a more balanced presence. Their appendices contained more information and practice of grammar and vocabulary items (glossary, phrasal verbs, prepositions) and a writing guide.
The coursebooks’ content suppling the instructional material used by the teachers in the observed sessions covered the following sections: in Textbook I, vocabulary (which included listening activities too), pronunciation and grammar (with a speaking practice activity as well); reading in Textbook II and reading and grammar in Textbook III. The reading sequences in the second and third coursebooks included vocabulary and speaking activities too. The topics of the units to which the activities belonged were money, music and traditions (respectively Textbooks I, II and III).
4.4. Classroom Observation
The purpose of this exploratory multiple case study was to observe actual use of materials rather than relying on teachers’ perceptions of their classroom practices when using textbooks, which might not be aligned with their classroom practices (Menkabu & Harwood, 2014). Accordingly, in parallel to Schwab (Schwab, 2022), the method of data collection selected was naturalistic classroom observation. Such classroom observations were conducted in lessons officially scheduled in the educational centers where they took place. This measure prevented data collection from interfering with the natural processes of instruction and learning (Borg, 2013).
The observed sessions were video-recorded to allow for a more comprehensive and accurate description of the data. Prior to the video-recording, I obtained consent for same from the Heads and the participant teachers of each educational center, following the procedure that the Heads had informed me about when I initially contacted them. This procedure also included the teachers informing the Upper-secondary-school students’ parents about the video-recording and collecting the corresponding consent forms. In the Official School of Languages, there were no students under eighteen years of age and they all gave their consent for the video-recording.
In order to mitigate the potentially undesirable reaction of the alteration of the teachers’ and students’ natural behavior or the “observer’s paradox” (Labov, 1972, as cited in Friedman, 2012), I adopted the following measures. Firstly, I carefully explained to the Head and to the potential participating teachers at a first and a second meeting respectively that the focus of the study was not evaluative but descriptive. I also reassured the people involved at the different meetings and the students in the classrooms that the objective of the video-recording was to obtain data about the natural processes occurring in the teaching sessions, instead of measuring students’ learning gains or evaluating the teachers’ performance. However, I tried not to disclose to the teachers explicitly that my objective was to analyze how they used the textbook, as I wanted to avoid the risk of altering their natural behavior when they taught. Secondly, I guaranteed the anonymous treatment of all the data collected to the Heads, the teachers and their students. Thirdly, one session in each center was recorded prior to the three officially assigned sessions that would provide the data for analysis so that the participants would become familiar with the procedure. Fourthly, the person in charge of video-recording (an acquaintance of mine who had ample experience in video-recording) placed the digital video-camera at the back of the classrooms in every single lesson observed (Humphries & Gebhard, 2012) and I sat down in the same position. I also informed all the participants and the Heads of the centers that the role of this person was merely technical and that he could not speak English.
The data that emerged from the classroom observations consisted of extensive field notes taken by myself and the transcriptions of four entire sessions: one randomly selected preparatory session that would be used as practice for the different coding stages and those three that provided the data for analysis. The preparatory sessions were video-recorded one week before the other ones. A research assistant transcribed the content of the four sessions verbatim. The transcriptions included the teachers’ and the students’ utterances. I decided not to take part in this methodological stage to avoid a learning effect that could have affected my perceptions in the different inter-rater analyses performed later (see the sections called Analysis of the Adaptation Techniques in Textbook Use and Analysis of the Measurement of FFI and MFI). In other words, I intended to remain in the same tabula-rasa situation as the second rater. The total amount of video-recorded minutes amounted to 213, distributed as follows: 55.5 (class of teacher 2), 70.5 (class of Teacher 3) and 87 (class of Teacher 1).
4.5. Data Analysis
4.5.1. Statistical Analyses
The descriptive statistical analysis of the sample relied on basic descriptive methods. In the case of the qualitative variables, the number of cases present in each category and the corresponding percentage were obtained. For the quantitative variables, the minimum and maximum values, the mean and the standard deviation (SD) or median and interquartile range (IQR) values were computed. Once the assumptions of normality (Shapiro-Wilk test) and homogeneity of variances (Levene test) were verified, the comparison of quantitative variables between groups was performed with the ANOVA test, while the Chi-square test or Fisher's exact test was used for the qualitative variables in the comparison between groups. To determine the effect of activity, center and the interaction of both factors on the FFI score, a two-factor ANOVA model was carried out through the General Linear Model (GLM) procedure. The degree of agreement in the qualitative variables was measured with the kappa coefficient (Cohen, 1988) and with intra-class correlation for quantitative variables (Fleiss, 1986). The statistical analyses were executed with IBM SPSS Statistics for Windows, version 26 (IBM Corp., Armonk, N.Y., USA). The differences considered statistically significant were those whosep < 0.05.
4.5.2. Analysis of the Adaptation Techniques in Textbook Use
As reported in the section called Methodological Considerations, the studies which reported quantitative data had not undertaken any inter-rater reliability measures. For the coding and analysis of the adaptation techniques in the present study, three inter-rater analyses were performed by myself as the first rater and another specialist in FLT research as the second one.
First of all, the number of activities in each one of the three lessons had to be determined. During my observation in situ of the sessions, I noticed that the teachers did not always follow the textbooks’ activities and/or sequence. Accordingly, the initial type of inter-rater reliability concerned the segmentation or establishment of the boundaries between the different activities conducted by the teachers. For that purpose, the second rater and I carefully watched the randomly selected session out of the three preparatory ones that were used to make the participants familiarized with the video-recording and procedures (see the Classroom Observation section). This second rater and I also had access to the transcription of the session and its corresponding textbook pages. Both of us took notes of the number of the different activities that we had detected and discussed those cases where our divisions did not coincide (given the ambiguous nature of certain boundaries resulting from some teachers’ adaptations) and we finally reached a consensus on how to differentiate between these potentially less clear cases. Afterwards, we each watched the video-recordings separately of the three sessions that provided the data for the analysis and noted the number of activities in each one. I identified 11, 8 and 13 activities in Educational Center A, B and C respectively, whereas the second judge noted 10, 8 and 13. The kappa value was 0.987 (95% CI: 0.915-1), which suggests an excellent degree of agreement between both judges (Cohen, 1988). The total number of activities implemented by the teachers was fixed in 32. The second rater and I considered that the beginning of an activity conducted by a teacher was marked with an instruction uttered by her which explicitly referred to such a start (e.g., “OK, let’s move on to activity number 5”). In turn, this clearly showed the end of the previous activity. In the case of the textbooks’ activities, the distinction was straightforward as we relied on the typographical divisions established by their corresponding authors.
The second step in the analysis of the teachers’ use of the prescribed textbooks consisted of determining whether the activities implemented by them were adapted or whether they complied with the textbooks’ original content. Accordingly, a second inter-rater reliability analysis was performed, for which purpose I undertook the same procedure as in the first one. When re-watching the preparatory session, the other judge and I noted all our perceptions regarding the existence of any adaptation technique or not, discussed our impressions and conflictive cases and reached a consensus. Afterwards, we each watched and analyzed the video-recording separately of the three sessions. The kappa value was 0.939 (95% CI: 0.857-1), which reflects an excellent degree of agreement between both judges (Cohen, 1988).
The third step implemented for the analysis of how the teachers used the textbooks involved the identification of the different adaptation techniques undertaken by them. I planned the corresponding inter-rater reliability analysis, whose procedure was the same as in the first and second steps. After taking notes and discussing some conflicting cases of the preparatory session, the second rater and I re-watched it again and we reached a consensus on the coding scheme to analyze the adaptation techniques. It should be taken into account that although both raters were knowledgeable about the specialized literature reviewed in this article, I considered it more appropriate to approach the analysis from a bottom-up perspective without any preconceived categories (Nguyen et al., 2018). In this rehearsal meeting, my colleague and I, as researchers, recurrently examined both the clearest and most conflictive cases. Such a reiterative analysis led me to adopt the following coding decision: the distinction between internal and external types of adaptations. The former referred to changes within a single activity; that is, either an alteration of its content (addition, removal or replacement of some items) or changes in its procedure, such as the reordering of its elements, altering classroom dynamics (e.g., modifying the textbooks’ originally intended interactive patterns from students’ exchanges to teacher-students’ exchanges), changing the modality of input (from written to aural and vice versa), etc. External adaptations entailed modifications throughout a unit of work since they covered the addition and omission of entire activities, resequencing or altering the original position of an activity, and merging activities into a single one (a category absent in previous literature).
Table 2 shows the coding scheme and the correspondence between the categories devised for the purposes of this study with the terminology of the previous authors in Table 1. As can be observed, the data did not reveal any instance of exploitation, rewriting, simplification, complexification or differentiation (McGrath, 2013, 2016). The category of extemporization was constantly used by the teachers. Thus, following Schwab (Schwab, 2022), this category was not included in Table 6 and Table 7 of the Results section. Moreover, the categories were not mutually exclusive (except for omission and addition of activities). The section called Examples of Analyzed Activities: Coding of Adaptation Techniques and Measurement of FFI and MFI includes examples of all the adaptation techniques from Table 2 that were identified in the data of this study, together with the values of FFI and MFI in each activity resulting from the application of the scale.
![]()
Table 2. Coding framework of adaptation techniques (correspondence between the terminology of the previous authors in Table 1 and that used in the current study).
Similarly to the first and second inter-rater analyses, the second judge and I watched and analyzed the video-recording separately of the three sessions following the coding scheme of Table 2. The corresponding kappa values for each adaptation technique are included in Table 3. As can be seen, the degree of agreement between both of us ranged from moderate to strong (Cohen, 1988).
4.5.3. Analysis of the Measurement of FFI and MFI
FFI and MFI in the textbooks’ activities and their corresponding adaptations and unchanged cases were measured with a scale purposefully designed for this study. In order to devise the different items that the scale comprised, I consulted the previous literature, which included an initial attempt to measure explicit and implicit teaching (Criado et al., 2010) as well as FFI and MFI (Criado, 2016) in textbook activities. I did not find any scales explicitly devised to measure FFI and MFI in the context of in situ or classroom textbook consumption. As stated in the Patterns in Classroom Textbook Use section, to analyze both textbook tasks and those implemented in classrooms Dao and Newton (Dao & Newton, 2021) drew upon Littlewood’s (Littlewood, 2004) continuum from focus on forms to focus on meaning. It highlighted non-communicative learning, pre-communicative practice, communicative language practice, structured communication and authentic communication. As Littlewood stated, “the categories shade into each other, and five is merely a convenient number” (p. 323).
![]()
Table 3. Inter-rater reliability (kappa) for the identification of the adaptations applied in the activities: general types (internal and external) and specific techniques.
The following description largely relies on this author’s own account (pp. 322-323). Basically, non-communicative learning focuses on the structures of the language (their formal properties and meaning); for instance, substitution exercises, discovery and awareness-raising activities. Pre-communicative activities could be assimilated to meaningful practice (such as question-and-answer exercises), which is not mechanical but it still does not entail filling an information gap. New information is conveyed in communicative language practice with pre-taught language in information-gap or personalized activities, whereas structured communication, as in structured role-play and simple problem solving, entails some degree of unpredictability. Authentic communication involves using language to convey unpredictable messages.
Despite certainly being an appealing option due to its clarity and comprehensiveness of its underlying constructs, I decided not to use Littlewood’s (Littlewood, 2004) scale for the following reasons. Firstly, it entailed the ascription of each activity to just one category, which means that attention was placed on the type of activity instead of on measuring FFI and MFI within the same activity. Secondly, the wording of the categories did not seem to include receptive-skill based activities. Importantly, to be fair to Littlewood, this author had not explicitly stated that his categorization intended to either gauge FFI and MFI, or that it could be used both for textbook content and use analyses. Thus, for the purposes of this study, I considered it necessary to devise a scale that could potentially capture the presence of FFI and MFI (whether mixed or not in the same activity) both in the context of coursebook content and consumption. As a starting point in the design of this scale, I resorted to Littlewood’s (Littlewood, 2004) proposal. A reviewer drew my attention to the relevance of Spada and Lightbown’s (Spada & Lightbown, 2008) concepts of isolated and integrated form-focused instruction for the scale. However, the crucial feature of timing of implementation in the integrated variant made it very difficult to include it in an instrument targeted at both textbook content analysis and use. The embedding of form-focused activities while the students are engaged in communicative tasks refers to synchronous teaching actions, which cannot be included or worded as a specific feature of a scale applied to textbook content analysis (similarly to reactive FFI or corrective feedback).
Figure 1 reflects the scale to measure FFI and MFI in the activities, both in their original shape from textbooks and as used by teachers. This final version emerged after my own reflection on Littlewood’s (Littlewood, 2004) proposal and a round with experts in the FLT field who contributed several slight, though worthy changes. The items are worded in such a way that they reflect the characteristics attributed to FFI. The presence of FFI in each item (if applicable) should be marked with the 1 number for FFI and with 0 for MFI, in accordance with their complementary relationship in a unidimensional variable (instruction). The total value out of 4 points is then converted to a 10-point scale. When analyzing the activities, I resorted to a spreadsheet file where I had added the pertinent formulas. Given the
![]()
Figure 1. The scale to measure FFI and MFI in textbook and teacher-implemented activities.
aforementioned complementary relationship between FFI and MFI, Table 8 in the Results section will only report the values of FFI. For examples of activities analyzed with this scale, see the next section (Examples of Analyzed Activities: Coding of Adaptation Techniques and Measurement of FFI and MFI).
“Focuses on analyzing language elements” somewhat parallels Littlewood’s (Littlewood, 2004) category of “non-communicative learning”, regardless of the level of metalanguage required (DeKeyser, 2009). I decided to include the explicit presentation of both structural and lexical features in this first item (whether deductively or inductively performed), as well as controlled input-based practice activities which require the students to focus on certain lexical forms (such as choosing the correct alternative to fill in the gap in a sentence). Also, receptive-skill activities are covered when the corresponding MFI cell of “Aims for controlled output use of the language” is marked with 1. “Favors linguistic accuracy” can be reflected not only in rule presentation or pronunciation drills, for instance, but also in meaningful output-based practice (similar to Littlewood’s pre-communicative practice). The presence of corrective feedback in the scale can be assimilated in the features “Favors linguistic accuracy” and “Focuses on analyzing language elements” (the latter depending on the degree of explicitness of the feedback strategy). For instance, if a teacher recurrently corrects certain students’ ill-formed responses when checking the answers to the questions of a reading skill activity, such an activity would probably not reflect a 100% use of MFI (despite its primary original emphasis on meaning). An activity can aim at making students produce controlled output but it can either favor linguistic accuracy (as in a meaningful drill) or pursue a balance between accuracy and fluency, prompted by the existence of a communicative gap (as is the case of communicative drills, or Littlewood’s communicative language practice and structured communication). The lack of a communicative gap in the fourth item refers to the unpredictable nature of real-life communication highlighted by Littlewood. Accordingly, this gap should not be reduced to lexical meanings but it should refer to communicative messages, present in either focused tasks or purely communicative activities (see the section called Form-focused and Meaning-focused Instruction [FFI and MFI]).
Finally, similarly to the coding of the adaptation techniques, an inter-rater reliability analysis was implemented for the application of the scale to measure FFI and MFI in each one of the 66 activities identified in the data (see Table 4 in the Results section). Again, the second rater and I watched the same preparatory lesson used to familiarize ourselves with the adaptation techniques, we applied the scale together, discussed the conflicting cases, reapplied the scale as often as necessary when doubts arose and reached a consensus on the less clear cases. Afterwards, both of us applied the scale to all the 66 activities on our own. The coefficient of the intra-class correlation was strong (Fleiss, 1986): 0.832 (95% CI: 0.829-0.841).
4.5.4. Examples of Analyzed Activities: Coding of Adaptation Techniques and FFI and MFI
This section offers nine examples of analyzed activities. The first eight are related to the adaptation techniques reported in Table 2 and the last one includes an unchanged activity.
Example 1. Changing the procedure (class of Teacher 3).
The instructions of the textbook’s activity asked students to read the first sentence of each paragraph of a text and to identify which paragraphs discussed several listed topics. The topic was “crickets for dinner”. The teacher adapted the original activity by changing its procedure. She told her students that she would read the text aloud and asked them to read it at the same time as her. She also explicitly told them to listen carefully to her pronunciation of the full text. She stopped at specific words, repeated them, and asked students to repeat them chorally after her. After each paragraph, she prompted her class to give her the answers following the instructions of the textbook’s original activity. Therefore, she transformed this activity into a mixture of a reading comprehension activity and a listening activity with an explicit focus on pronunciation. The values of FFI and MFI are included in Figure 2. As can be seen, if the activity had not been adapted, the overall points of MFI would have reached 10. However, the teacher’s direct emphasis on pronunciation implied a focus on linguistic accuracy and hence the change of the MFI value to 7.5 points.
Example 2.Changing the procedure (class of Teacher 2).
This activity is the first one out of the six that integrated a reading sequence in the textbook. The text focused on the topic of the power of music to change people’s lives in very difficult circumstances. Given that Examples 2, 3, 4 and 5 will deal with the same sequence, a brief description of which will be given now. The instructions of the textbook’s first original activity asked students to guess why music is often called the universal language. It was followed by two more activities centered on processing the meaning of the text by means of skimming it and locating details respectively. The fourth activity asked students to complete sentences from the text with their own words. The fifth activity focused on details of the content again and, finally, the sixth one required students to find words or expressions in the text that meant the same as the list provided in the instructions.
The teacher modified the interactive patterns of the first textbook’s activity. Rather than getting students to talk in pairs, she asked the questions in a choral way to her class and triggered individual students’ responses, which were often restricted to a short sentence (e.g., “Because everybody likes music”). The values of FFI and MFI were the same as those of the textbook’s original activity (see Figure 3). The adapted activity was still meaning-focused as it could be considered that it entailed speaking practice for the few students asked by the teacher, while it represented a listening activity for all the other ones. This specific materialization of the “changing the procedure” adaptation technique was recurrently performed by the three teachers.
Example 3.Adding an activity (class of Teacher 2).
The teacher added an activity she herself created in the reading skill-based sequence to which Example 2 belonged. The first activity was aimed at activating the students’ background knowledge and motivation (see Example 2). Afterwards, the teacher showed images of music groups to the students (Abba, Queen…) and asked them whether they were familiar with such groups. Individual students responded. She briefly explained the story behind such groups (their countries of origin, achievements, etc.) and played some brief extracts of three songs. This added activity entailed listening practice focused on the communicative messages
conveyed by the teacher: the students had to pay attention to her content explanation about the music groups, which could be considered to represent an information gap (regardless of how much the students knew about such groups). Accordingly, as can be seen in Figure 4, the value of MFI totaled 10 points.
Example 4.Reducing content (class of Teacher 2).
The targeted textbook’s original activity was the fifth one in the reading sequence introduced in Example 2. It consisted of choosing the right answer out of four in multiple-choice items related to the content of the text. The teacher asked her students to answer the first three items on their own. Both the third item and the one which was omitted required a higher degree of inference than the first two. She did not either replace or add any other item to compensate for the fourth one. The values of FFI and MFI in the textbook’s original activity were not affected by the teacher’s adaptation, given the strong meaning-focused nature of the activity (see Figure 5).
Example 5.Resequencing an activity and adding content (class of Teacher 2).
The targeted textbook’s original activity was the last one of the six-activity reading sequence in which Examples 2, 3 and 4 are located. This activity required the students to find five words or expressions in the text that meant the words and phrases listed next to the instructions. Consequently, the students had to rely on the co-text and context of such items. The teacher added two
more words (“soaring” and “point out”). She set this activity after the addition of the one she had created (Example 3). Following the lexical activity object of this Example 5, she instructed her students to complete the textbook’s fifth original activity, which presented four reading comprehension items in a multiple-choice format (Example 4). Then, she conducted the third activity from the textbook (sentences about which the students had to decide whether they were true or false plus give evidence from the text in the latter case), followed by the textbook’s second original activity (focused on identifying the author’s purpose in the text or general idea). The teacher omitted the textbook’s fourth original activity, which asked students to complete five sentences from the text with their own words.
Thus, as can be seen, by resequencing the sixth and final activity in the textbook’s original sequence as the third one in her own adapted one, the teacher prioritized lexical understanding before reading comprehension per se. The FFI and MFI values of the textbook’s sixth original activity were not affected by this teacher’s adaptation (see Figure 6).
Example 6.Replacing content (class of Teacher 1).
This activity entirely focused on forms practiced the pronunciation of the letter “o” by saying some sentences. It is the last one in a sequence of six activities where the students had: classified words according to the three different possible sounds of this letter, analyzed how the letters “or” were normally pronounced when they were stressed and listened to the textbook’s audio material to check all their answers. The teacher replaced one sentence with another one (“the police caught the thieves”), which did not alter the FFI and MFI values of the textbooks’ original activity (see Figure 7).
Example 7.Omitting an activity (class of Teacher 1).
In the three observed sessions, the types of omitted activities were varied. They ranged from meaning-focused cases such as a scanning activity for reading, and semi-controlled activities asking students to complete sentences from a text using their own words to form-focused exercises that required the students to remember sentences containing specific vocabulary and previously practiced pronunciation.
Example 7 belongs to this last strand of activities. The targeted activity followed vocabulary practice about prepositions in sentences. The activity omitted by the teacher asked students to remember the prepositions. As can be seen in Figure 8, the values of FFI in this activity totaled 10 points.
Example 8.Merging and resequencing activities (class of Teacher 3).
The targeted activity was an explanatory metalinguistic note about modals in conditionals, informing the students about the fact that “can”, “may”, “might” and “should” frequently replace “will” in the first conditional, while “could” and “might” can replace “would” in the second one (e.g., “If you want, we can go the cinema this evening”). A sample sentence was also offered. This metalinguistic note was the fourth activity in the section devoted to grammar in the textbook’s unit (first and second conditionals). The two first activities represented an inductive presentation of such structures. The teacher merged the metalinguistic note with the first activity, thus resequencing it as well. The values of FFI and MFI were not altered by the teacher’s adaptation, given the strong form-focused nature of the metalinguistic note (see Figure 9).
Example 9.Unchanged activity (class of Teacher 3).
Many unchanged activities in the three observed sessions concerned the explanation and practice of grammatical structures and vocabulary. Example 9 deals with a communicative drill, which is very frequently used in textbooks. It was located in a grammar section about the present perfect and past simple. After the students had filled in the gaps of a text with the right tense, they engaged in inductive and deductive presentation activities as well as controlled practice exercises. The instructions of the communicative drill directed students to work in pairs and interview one another with eight prompts about the topic of the text previously read (spending money); for instance, “Have you ever lent money to someone who didn’t pay you back?” The questions required the students to practice the present perfect and the simple past tense. However, the nature of the prompts encouraged them to answer following their own experiences and opinions, which constituted information gaps that the interviewers filled in with the content provided by the interviewees and vice versa. As can be observed in Figure 10, the activity presented a balanced proportion of FFI and MFI.
5. Results
Before focusing on each one of the research questions, a general account of the activities will be provided. Table 4 includes the descriptive data of all the 66 activities analyzed, which comprised 34 textbooks’ original activities and 32 implemented by the teachers.
The mean time of all the classroom activities was 5.6 ± 5.2 (session of Teacher 1); 6.9 ± 3.4 (session of Teacher 2) and 5.9 ± 2.6 (session of Teacher 3). This
![]()
Table 4. Activities from the textbook and implemented by the teachers.
mean time included both the time that the teachers spent on setting up the activity before the students completed it (explaining instructions, modelling examples, etc.), the time of the students’ performance and the time needed for correction in class. No statistically significant differences were observed between the classes (F(2;29) = 0.3, p = 0.744).
The first research question asked to what extent there were any differences in the percentage of adapted activities when instructors were teaching three different classes of the same linguistic level. Concerning the textbooks, 10 activities out of the total analyzed 34 were omitted (29.4%). As can be seen in Table 5, no statistically significant differences in the percentage of textbooks’ omitted activities were observed between the teachers. Regarding the activities implemented by the teachers, 24 out of the total 32 activities were adapted (75%). There were no statistically significant differences between the teachers either, as can be seen in Table 5 too.
Table 6 shows the descriptive information about the specific adaptation techniques applied by the teachers. The most frequent techniques were changing the procedure and adding activities (internal and external types of adaptation respectively).
The second research question asked to what extent there were any differences in the percentage of the adaptation techniques when instructors were teaching three different classes of the same linguistic level. As can be observed in Table 7, there were no statistically significant differences between the teachers’ classes.
Finally, the third research question inquired to what extent there were any differences in FFI and MFI between the textbooks’ original activities and those implemented by the instructors while teaching three different classes of the same linguistic level. No statistically significant differences were found (Table 8).
The quantitative results can be summarized as follows:
1) No statistically significant differences between the three teachers were observed concerning the percentage of the omitted and non-omitted activities (Table 5) and of the adaptation techniques they applied (Table 7).
2) The comparison of FFI and MFI in the textbooks’ content between the three coursebooks yielded non-statistically significant differences too (Table 8).
3) The same pattern was found in FFI and MFI according to the teachers’ textbook use. In other words, the type of instruction delivered in the three observed sessions was the same (Table 8).
![]()
Table 5. Omitted textbooks’ activities and activities conducted by each teacher: descriptive and comparative data.
Note: as indicated in the section called Analysis of the Adaptation Techniques in Textbook Use, extemporization was used by the teachers in their arrangement of virtually every single activity. Therefore, no percentages were computed.
![]()
Table 7. Adaptation techniques in the activities implemented by the teachers (n = 32).
![]()
Table 8. FFI: descriptive data and statistical contrasts of the activities analyzed (n = 66).
d.f.: degrees of freedom. TE: typical error. η2: Partial eta squared (effect size).
6. Discussion
Three essential preliminary notes should be considered before undertaking the discussion of the previous results:
1) The reader is reminded of the abundant methodological differences among the studies as explained in the Methodological Considerations section.
2) In the comparison of my own results with those of the previous literature which reported quantitative data, the following aspects should be taken into account. Firstly, I computed the means of the percentages provided by Loh and Renandya (Loh & Renandya, 2015) and the percentages of the frequencies included in other studies to obtain their corresponding means (Dao & Newton, 2021; Marcos Miguel, 2015, Nguyen et al., 2018 and Schwab, 2022. In the case of Marcos Miguel, I inferred the frequencies from her figure 1 on page 315). Secondly, as stated in the fifth point of the Methodological Considerations section, only the frequencies of the added activities in Dao and Newton (Dao & Newton, 2021), Marcos Miguel (Marcos Miguel, 2015) and Schwab (Schwab, 2022) can be considered (without percentages). Thirdly, regarding Rathert and Cabaroğlu (Rathert & Cabaroğlu, 2021), percentages could not be computed as the frequencies reported by these authors only referred to the number of the adaptation techniques performed, with no information about the total number of activities implemented in the classroom (see the fifth point in the Methodological Considerations section).
3) None of the quantitative studies mentioned in the section called Empirical Studies on Classroom Textbook Use reported inferential statistics about the percentages of the adaptation techniques performed by the teachers, an aspect that also affects the few studies which explicitly provided frequencies or percentages of form-focused and meaning-focused activities (Dao & Newton, 2021; Seferaj, 2014; Tran et al., 2023). Thus, regarding both elements—adaptation techniques and types of instruction—the nature of the comparison between the quantitative results of the previous studies with those of the present one will be descriptive and global.
The first research question addressed the extent to which there were any differences in the percentage of adaptations conducted by the teachers in the three observed sessions. As stated in the Results section, the teachers reflected a homogenous pattern regardless of the educational stage they taught at (Upper-secondary education versus the non-university special regime to which Official Schools of Languages belonged). Perhaps this finding was due to the teachers’ similar profile in terms of age and years of experience, as well as the environment (Spanish EFL in formal contexts), which is heavily constrained by the regulations in terms of the curriculum to be covered. However, these reasons will remain speculative until further research in other Spanish EFL B1 classes is conducted. The percentage of adapted activities conducted in the classrooms was high: 75%. For the purposes of comparison with other previous studies, let us remember that, out of this 75%, 66.6% (16 activities) did not correspond to added activities (see Table 5 and Table 6). All the studies previously reviewed revealed lower percentages of non-added activities, ranging from 13.3% in Schwab (Schwab, 2022), 24.8% in Marcos Miguel (Marcos Miguel, 2015), 39.4% in Loh and Renandya (Loh & Renandya, 2015) to 57.1% in Nguyen et al. (Nguyen et al., 2018). In general, the result of this study converges with a generalized pattern of teachers’ classroom adaptation in the literature available, including the qualitative studies, despite the higher or lower emphasis on adaptation (Grammatosi & Harwood, 2014; Li & Xu, 2020; Menkabu & Harwood, 2014; Seferaj, 2014; Shawer, 2017; Tasseron, 2017; cf. Aftab, 2022).
The results of the second research question revealed no statistically significant differences between teachers in the percentage of the different adaptation techniques undertaken. They omitted 10 out of the 34 textbooks’ original activities (29.4%), a percentage which is close to Dao and Newton’s (Dao and Newton, 2021) and Schwab’s (Schwab, 2022) results (23.9% and 34.8% respectively). Still, it is very far away from the 4.8% of the minus category reported in 2015 by Loh and Renandya (which also included omission of content, following Tomlinson & Masuhara’s [ Tomlinson & Masuhara, 2004] classification), as well as from the other extreme represented by Abdel Latif (Abdel Latif, 2017), with almost half of the activities omitted, and Marcos Miguel (Marcos Miguel, 2015), who reported 65.6% of omitted activities. The prioritization of form-focused work due to exam pressure and time restrictions seemed to account for the deletion of other meaning-based activities in these studies, an explanation that concurs with the reasons stated by the teachers of the current one in the informal conversations I held with them after the classroom observation had taken place.
The next most recurrent type of adaptation in this study was changing the procedure (28.1%). Modifying was also the second most frequent adaptation technique in Marcos Miguel (Marcos Miguel, 2015) and Schwab (Schwab, 2022), representing 19.4% and 9% respectively. Such results emphasize the prevalence of altering the task modality of activities regardless of the status of the language targeted (foreign in the case of Schwab and this study and domestic in Marcos Miguel). As stated in Example 2 of the section called Examples of Analyzed Activities: Coding of Adaptation Techniques and Measurement of FFI and MFI, a common occurrence in the sessions observed in this study was the transformation from a student-student pattern of interaction to a teacher-fronted one (following Sinclair & Coulthard’s [ Sinclair & Coulthard, 1975] traditional Initiation-Response-Follow-up/IRF sequence), which coincides with Dao and Newton’s (Dao & Newton, 2021) and Seferaj’s (Seferaj, 2014) observations. As the latter asserted, this modification both distorted the instructions of the textbooks and reduced the possibilities of meaningful interaction between students. However, as also indicated in the aforementioned Example 2, and as an advancement of the discussion of the third research question, the values of FFI and MFI in such activities implemented by the teachers were not affected by the textbooks’ original activities. The instructors’ interventions were still meaning-focused: except for the few students who briefly answered their teachers’ questions, for the remaining ones the adaptation undertaken paralleled a listening activity.
After omission of activities and changing the procedure, adding activities was the third most frequent adaptation technique in the present study. It represented 25%, a result considerably lower than the 67.2% reported by Ngyuen et al. (2018). Perhaps the observed difference can be attributed to the fact that the type of activities exclusively involved oral tasks in the latter. Contrary to Seferaj (Seferaj, 2014), the teachers in this study did not extensively supplement their pedagogical material with extra grammar work (which was already included in the textbooks’ last pages), but they mostly added cultural discussions/presentations such as in Example 3 (see the section called Examples of Analyzed Activities: Coding of Adaptation Techniques and Measurement of FFI and MFI). Similar to most instructors in Schwab (Schwab, 2022), the teachers constantly resorted to extemporization (cf. Rathert & Cabaroğlu [ Rathert & Cabaroğlu, 2021], who focused on pre-intermediate and intermediate textbooks). Extemporization was materialized in this study as the explanation and paraphrasing of instructions in L2 English, or the translation and explanation of instructions into L1 Spanish (in the case of the sessions from the secondary schools), or modelling examples before the students did the activities on their own; a far less common variant was Schwab’s resorting to one or two learners for demonstration prior to the students’ completion of the activities. The conjunction of extemporization with the 18.8% revealed in adding content, which represented the fourth most frequent adaptation technique, appears to point to the scaffolding nature of the teachers’ pedagogical profile.
Resequencing of activities was the fifth most frequent adaptation technique (9.4%). Its less common presence is a finding that aligns with Marcos Miguel (Marcos Miguel, 2015) and Schwab (Schwab, 2022), where this technique—called “reordering” in both studies—was the least frequent one (5.4% and 4.3% respectively). It seems that the educational stage targeted (Primary Education in Schwab, Higher Education in Marcos Miguel and Upper-secondary Education and non-university special regime teaching in this study) did not affect the results, though much more research is needed to confirm or refute this pattern. Reducing and replacing content were used in the same proportion as resequencing activities (9.4%). Finally, another non-frequent technique that emerged in the data and which had not been mentioned in previous literature was merging activities (6.3%). The two examples found in the present study referred to the merging of explanations into a single activity. This allegedly unveils a practical, pedagogically based strategy applied by Teachers 2 and 3, with the plausible aim of providing a more comprehensive and complete explanation of all the nuances of the targeted grammatical structures (rather than dividing the students’ attention in two different time periods).
In terms of Shawer’s (Shawer, 2010, 2017) textbook-user profiles, it is obvious that traces of the coursebook-maker role were absent in the data. Shawer (Shawer, 2017) indicated that the percentages of adaptation implemented by the teachers whose classes he observed were 0% for the textbook transmitters, 62% for the textbook developers and 95% for the textbook makers. Interestingly, in the current study the percentage of omitted activities was 29.4%, while that of the adapted activities conducted in the classrooms was 75%, with no statistically significant differences between teachers in either case. Following Shawer’s results, does the result in this study mean that the three teachers could be assigned to a middle point between textbook developers and makers?
Certainly, the teachers in this study adapted activities considerably, but the nature of their adaptations did not entail a deep and critical change in terms of the textbooks’ original activities. Exploitation and simplification, for instance, did not appear in the data. The teachers did not resequence activities, reduce or replace content frequently. Moreover, although changing the procedure was the second most recurrent adaptation technique after the omission of activities, it mostly consisted of the teachers distorting the original purpose of speaking activities in textbooks by fostering a teacher-fronted pattern rather than proper interactive practice (see Example 2 in the section called Examples of Analyzed Activities: Coding of Adaptation Techniques and Measurement of FFI and MFI). However, while I observed the sessions, I could perceive that the three teachers tried to render learning more accessible to their students by recurrently resorting to extemporization (see the Analysis of the Adaptation Techniques in Textbook Use section) and by adding extra items—especially in the presentation and practice of language features—that could reinforce the students’ development of linguistic knowledge. Likewise, in the informal conversations that I held with the teachers after each observation, the three of them told me that they somehow attempted to motivate students by adding activities in the shape of discussions about topics of the unit dealt with. Thus, even though the teachers in this study did not reach the status of textbook developers (given their important adherence to the textbooks’ content and sequence of activities), it could be argued that they adopted a somewhat “advanced” or more “sophisticated” textbook-transmitter role, neither a “master” nor a full “slave” (in Rathert & Cabaroğlu’s [ Rathert & Cabaroğlu, 2021] terms). Overall, the nature of their textbook use seemed to be context-bound (Grammatosi & Harwood, 2014; Rathert & Cabaroğlu, 2021). The results also suggested that the boundaries between the teachers’ profiles of textbook use are to be understood in flexible terms and that a more precise account of teachers’ patterns of coursebook use requires the careful joint consideration of both qualitative and quantitative perspectives when examining the data.
The third research question was focused on determining to what extent there were any differences in FFI and MFI between the textbooks’ original activities and those implemented by the instructors while teaching three different classes of the same linguistic level. Once again, results revealed no statistically significant differences between centers and activities from the textbook or implemented by the teachers. Teachers used the same amount of FFI in their classes as that which was present in their textbooks (44.7% versus 43.5% respectively, as can be seen in Table 8). Conversely, Seferaj (Seferaj, 2014) reported a lower higher percentage of FFI than MFI in the activities conducted by the teacher (39% versus 58%, consisting of 33% of activities that integrated real-life receptive and productive skills and 25% of other meaning-based activities such as reading and listening comprehension ones). As opposed to the present results and Seferaj’s, an emphasis on form-focused activities implemented by the teachers was found in other studies (most of which did not offer quantitative data such as Aftab, 2022, and Tasseron, 2017; cf.; Tran et al., 2023), as well as in Dao and Newton (Dao & Newton, 2021), who reported 61% and 82% of low-communicative value activities in their analysis of the textbook’s content and teachers’ classes respectively. Perhaps these differences could be accounted for by two factors: the dissimilar linguistic level (A2 in Dao & Newton versus B1 in this study) and the use of two divergent measurement instruments (the scale purposefully devised for this study and Dao and Newton’s resort to Littlewood’s [ Littlewood, 2004] five categories in the continuum between focus on forms and focus on meaning). Future research comparing the application of both instruments in classrooms of different levels could contribute to shedding more light on this issue. Despite essential variables to consider in classroom contexts such as students’ age and level and logistic institutional constraints, an excess of form-focused activities does not enable students to become satisfactory language users, given the absence of pedagogical measures to help them proceduralize and automatize knowledge with activities that gradually simulate real operating conditions (Criado, 2022). Since textbooks are the prevalent and most important resource in the teaching kit (Harwood, 2022; Li & Xu, 2020, Rathert & Cabaroğlu, 2021, etc.), the amount of FFI and MFI they provide has a decisive impact on the actual affordances or learning opportunities granted to learners. This conclusion is endorsed in the present study by the absence of any statistically significant differences between the values of FFI in the textbooks’ content and that of the teachers’ classroom activities. Finally, the fact that the same amount of time was devoted to the latter activities seems to agree with the results of the equality found in FFI as well as in the adaptation techniques.
This study has several limitations, the remediation of which could be brought about by the inclusion of fruitful future research possibilities on textbook use, both concerning the measurement of adaptation techniques and types of instruction. First of all, as in any case study, the results cannot be extrapolated due to the small number of participating teachers, as well as the specificity of the context being researched. Secondly, although I made every effort to reassure teachers and students about the anonymous nature of the data collected and the actions adopted to mitigate the effect of video-recording on the teachers’ and students’ behavior, such a possibility cannot be totally discarded. Thirdly, an analysis of the correspondence between the teachers’ lesson plans and their actual classroom teaching in situ could also be a valuable tool in order to complete teachers’ profiles in textbook consumption, prior to and during class time (both planned and spontaneous performances). Fourthly, stimulated recalls could provide useful information about the teachers’ perspectives and reasons for adaptation of specific episodes recorded in their classes. Fifthly, evidence of the students’ reactions to the coursebooks and how teachers use them, as well as of the effects of textbook consumption on students’ learning (Harwood, 2021, 2022), is particularly necessary to contextualize the results of FFI and MFI derived from textbook consumption as precisely as possible. Sixthly, although this case study attempted to reveal a detailed snapshot of the patterns of teachers’ textbook use in the three observed sessions, it would be interesting to observe to what extent such patterns would be maintained over time regardless of the type of content being taught (for instance, whether a grammar content would trigger different patterns from a reading or a culture session). Finally, as far as I am aware, the scale used in this study constitutes the first systematized tool to measure FFI and MFI within the same activity in the analysis of textbook content and textbook use. In this respect, the resulting challenges should not be overlooked. For instance, the scale did not allow for the capturing of different interactive speaking patterns whose omission reduced opportunities for oral productive skill-based practice, a frequent option adopted by the teachers in this study and the previous literature (Aftab, 2022; Dao & Newton, 2021; Seferaj, 2014, etc.).
7. Conclusion
The present multiple case study aimed to provide a detailed comprehensive account of the patterns of textbook use, both regarding adaptation techniques and the differences in FFI and MFI between textbook content and textbook use arising from the implementation of such techniques and the selective retention of the textbooks’ original activities. It was conducted in the previously non-researched context of Spanish EFL classrooms from Upper-secondary schools and an Official School of Languages.
Despite the methodological limitations stated in the previous section, two important implications can be extracted from this exploratory research: the importance of textbooks for classroom teaching and the fact that instructors’ pedagogical actions largely abide by the textbooks’ content, which in turn points to the crucial importance of rigorously devised coursebooks. Also, this study has hopefully contributed to broadening scholarship understanding of textbook use patterns in general and about the resulting FFI and MFI presence in particular—an area that had hardly received any previous attention. In turn, another potential contribution is raising much-needed awareness in both fields of ISLA and Teacher Education of the crucial importance of textbook use for students’ development of communicative competence. Finally, it is important to highlight the methodological innovations implemented to ensure the reliability of data measurement as much as possible for the following aspects: activity segmentation, identification of adaptation techniques, the values of FFI and MFI (with a purposefully devised scale) and the total time devoted to all the classroom activities in order to test whether this factor could affect the presence of specific adaptation techniques and of FFI and MFI. It is expected that such methodological procedures will prompt researchers to take similar actions in order to enhance their methodological rigueur in future textbook use studies.
Acknowledgements
Work funded by the Spanish Ministry of Economy and Competitiveness: Research Project FFI2013-44979-R. This funding is hereby very gratefully acknowledged. Sincere thanks are also due to the teachers and the students who participated and to the Heads of the centers where the data collection took place.