A Comparative Study of Intercultural Communicative Competence Representation in Local and International ELT Textbooks Used in Iran

Introduction

English has stopped being just a school subject—it is now a tool for living across cultures (Humphreys, 2024). That shift has pulled English Language Teaching (ELT) away from a narrow fixation on forms and toward the broader goal of intercultural communicative competence (ICC)—the knowledge, skills, attitudes, and critical awareness needed to navigate meanings with people who do not share one’s cultural background (Byram, 1997; Deardorff, 2006; Dervin, 2016; Gui, 2024; Kramsch, 2011). ICC means recognizing cultural differences, interpreting behaviors in context, and adjusting how we communicate. In most EFL contexts, textbooks are the main vehicle for this work: they don’t just sequence grammar; they curate cultural worlds and tacitly teach learners how to meet the ‘other’ (Cortazzi & Jin, 1999; Tomlinson, 2012). The increasing body of research examining the influence of culture within ELT textbooks provides a clear explanation of how content evaluation and comparative analysis significantly shape the development of these educational materials (Ghoorchaei et al., 2021). Their choices shape not only classroom talk but also teachers’ professional identities and routines (Tavakoli, 2021). Yet, how textbooks operationalize ICC is deeply local. In many EFL contexts, especially where English is rarely spoken outside class, textbooks become the de facto intercultural mediator.

In the Iranian context, this matters even more because learners typically encounter English through a tripartite textbook ecosystem: the national school series (Prospect in lower secondary and Vision in upper secondary), private-sector materials used in institutes (e.g., ILI New Elementary), and international commercial series (e.g., American English File, Top Notch) used to ‘globalize’ exposure. Multiple evaluations of the national textbooks report limited variety of cultural content and low alignment with ICC standards (Banaruee et al., 2023). Private-sector materials such as the ILI New Elementary series typically broaden exposure yet still underrepresent systematic integration of source (Iranian) culture with target and international cultures (Hosseinzadeh et al., 2022; Rashidi & Najafi, 2010). International series like American English File and Top Notch richly stage global settings and social practices, but their Western-centric tilt can weaken perceived relevance and limit space for learners’ own identities (Dehbozorgi et al., 2014; Köksal & Ulum, 2021; Morady Moghaddam & Tirnaz, 2023). Textbooks often treat culture as a curated spectacle—something to look at, not negotiate- offer imbalanced intercultural contacts, and rarely position learners to interpret, relate, and discover across cultures (Abid & Moalla, 2020, 2022). Meanwhile, teachers routinely ‘localize’ or adapt materials to bridge these gaps, but such work is uneven and not always supported by the design of the books themselves (Atai & Mazlum, 2013; Graves, 2022; Shin & Crandall, 2014; Toledo-Sandoval, 2020).

Against this backdrop, the present study offers a focused, elementary-level comparison across the three streams most Iranian beginners actually experience: Prospect 1–3 (national) (Alavimoghadam et al., 2013, 2014, 2015), ILI New Elementary (henceforth ILI NE) 1–3 (private) (Baleghizadeh, 2017), and American English File at beginner/elementary level (Starter & 1) (Latham-Koenig et al., 2019). To ensure fair comparability, we operationalize ‘elementary’ as CEFR A1–A2. In American English File (henceforth AEF), that band is covered by AEF Starter (A1) and AEF 1 (A2); AEF 2 leans into pre-intermediate (≈B1) and would overshoot the Prospect/ILI NE level in both linguistic load and task complexity. Using Byram’s (1997) ICC model (knowledge, interpreting/relating, discovery/interaction, attitudes, critical cultural awareness) and Cortazzi & Jin’s (1999) source–target–international culture triad, we examine what cultures are represented, how intercultural encounters are staged, and whether tasks genuinely cultivate ICC for elementary learners. However, it is worth noting that Byram's model is not without limits. Xu et al. (2025), in a major review, argue that traditional ICC frameworks assume face-to-face communication in stable cultural settings. The authors call for a ‘Hybrid ICC Model’ that adds digital literacy and awareness of power relations, noting that even beginners can start noticing these dimensions. While the Hybrid ICC Model's added dimensions — digital literacy and platform-mediated communication, in particular — are well suited to studying learners' online and multilingual interactions, they have limited applicability to the print-based, teacher-mediated elementary textbooks examined here, which do not yet stage digital or multi-platform communication for A1–A2 learners. Byram's (1997) model, despite the limitations Xu et al. (2025) identify for advanced and digitally mediated contexts, remains the most established and comparably applied framework for analyzing how cultural content and tasks are structured within printed course materials.

This study extends an active line of Iranian ICC comparison research. Prior work has usefully contrasted pairs of streams: national vs. international (Dehbozorgi et al., 2014; Morady Moghaddam & Tirnaz, 2023) or national vs. localized-institute materials (Hosseinzadeh et al., 2022; Rashidi & Najafi, 2010), but no study to date has brought the national (Prospect), private-institute (ILI NE), and international (AEF) streams into a single, same-level analytic frame. Unlike Morady Moghaddam and Tirnaz (2023), who examined the upper-secondary Vision series against Top Notch, the present study targets the elementary/lower-secondary Prospect series specifically, alongside a private-institute stream absent from prior three-way comparisons. By bringing these series into the same analytic frame, the study can identify overlaps, asymmetries, and blind spots that matter for policy and practice. Findings are intended to inform curriculum revision discussions in public schools, materials-selection decisions in institutes, and teacher education on mediating culture—helping the ELT education system support beginners to sustain a confident sense of local identity while learning to participate in wider, plural publics (Baker, 2015; Risager, 2007).

Literature Review

From Linguistic Competence to Intercultural Competence

Over the past three decades, English teaching has moved beyond a narrow focus on forms toward preparing learners to communicate appropriately and effectively across cultural boundaries. In the textbook literature, these outcomes are often framed through the ‘intercultural speaker’ ideal: someone able to draw on cultural knowledge, attitudes, and interactional skills to engage new people and contexts (Byram, 1997). Crucially, multiple analyses remind us that textbooks are not neutral containers of grammar: they are cultural artefacts that curate whose lives appear, which social practices are normalized, and what kinds of contact scenarios are staged for learners to rehearse. Byram’s framework remains influential because it translates ICC into five categories that can be applied consistently to pedagogical materials. Hoff (2020) notes that the model can imply a static, nation-linked view of culture and understate fluid identities and conflictual dimensions of intercultural encounters. Dervin (2016) and Holliday (2010) likewise reject treating individuals as representatives of homogeneous cultural groups and emphasize the roles of discourse, context, and power. Xu et al. (2025) extend this critique to digitally mediated interaction and power relations, dimensions not captured by the present print corpus.

A practical implication is that which cultures a book depicts matters less than how learners are asked to engage with them. At beginner levels, this means tasks that nudge learners to interpret and relate meanings and to try out safe, age-appropriate interactional routines. Scholars such as Byram and Wagner (2018) argue that intercultural competence must be integrated throughout language teaching, not treated as an advanced add-on. Extending this principle to the elementary band, we argue that A1–A2 instruction is a natural starting point for routine pragmatic acts, rather than a stage where such work must wait.

Across national and global textbooks, the recurrent problem is not simply whether culture appears, but whether it is developed through tasks requiring interpretation, interaction, or critical evaluation. Maghsoudi (2020) and Rezaee et al. (2024) identify this descriptive emphasis in Iranian state series. Furthermore, Bagheri Nevisi and Reisi (2024) found superficial interaction and little critical reflection in Prospect and Vision conversations, while Rezaee and Saleh (2025) found that Prospect’s cultural content did not promote purposeful reflection on alternative interpretations. Yook and Kim (2024) report a comparable gap in Korean textbooks despite explicit curricular aspirations, whereas Keles and Yazan (2020) show that global coursebooks remain selective in the cultures and communities they represent. Together, these studies locate the gap between curricular aspiration and textbook design in task depth rather than cultural presence alone.

Cultural Representation and Orientation in ELT Textbooks

Cortazzi and Jin’s (1999) distinction among source, target, and international cultures offers a practical lens for comparing which cultural referents textbooks make visible. Its categories, however, can suggest that cultures are discrete and nationally bounded. Holliday (2010) treats cultural boundaries as fluid and negotiable, while Dervin (2016) warns against reducing individuals to stable national groups; Risager (2007) similarly emphasizes transnational cultural flows. The categories are therefore used here as coding heuristics for textbook referents, not as claims that cultural identities are homogeneous. López-Jiménez and Sánchez-Torres (2021) emphasize self-reflection and critical cultural awareness, whereas Baker (2022) argues that language education must address fluid intercultural and transcultural communication. Both perspectives imply that a balanced distribution of cultural categories cannot, by itself, demonstrate ICC.

Comparative work on local and international series shows that the direction of representational imbalance varies by context. Prihatiningsih et al. (2021), for example, found strong source-culture dominance in Indonesian seventh-grade textbooks, whereas Köksal and Ulum (2021) identified inner-circle hegemony in Touchstone. Aljamrah and Awwad’s (2025) comparison likewise found Headway strongly Western-oriented, although the Jordanian Action Pack distributed cultural orientations more evenly. A different analytical framework led Sadeghzadegan et al. (2025) to find that both Vision and Touchstone favored big-C themes; because that study counted themes rather than ICC tasks, higher cultural frequency cannot be read as stronger intercultural engagement. Across these studies, neither source dominance, target dominance, nor numerical balance reliably predicts ICC depth. The stronger question is whether tasks require learners to interpret, compare, interact, and critically evaluate perspectives.

The Iranian Tri-Stream Ecosystem at the Elementary Band: Prospect, NE, and AEF

English education in Iran comprises a centrally administered school system alongside a parallel private-institute sector, while international commercial coursebooks circulate mainly through non-state instruction. Gholaminejad and Raeisi-Vanani (2021) document the policy and textbook constraints of mainstream schools; Iranmehr et al. (2024) situate those constraints within a localized, limited-exposure language-education policy. Within this landscape, the present corpus represents three material streams—national school books, private-institute materials, and international commercial series—without assuming that individual learners encounter all three. Banaruee et al.'s (2023) nationwide evaluation of Vision 1–3 found culture and communication standards implemented at a low level overall, with only modest improvement at Vision 3. At the lower-secondary band central to this study, recent analyses converge on limited reflective depth: Bagheri Nevisi and Reisi (2024) found that superficial interaction dominated Prospect conversations and that critical reflection received little attention, while Rezaee and Saleh (2025) found that Prospect provided only superficial opportunities for independent reflection. This convergence across two coding frameworks suggests a problem of task design, not merely low cultural variety. Dehbozorgi
et al. (2014), who coded pictures, dialogues, and speaking activities, found that lifestyle-oriented little-c content dominated AEF and that orientation estimates depended on the unit of analysis.

Other Iranian comparisons show that findings vary by level and analytical lens. Rezaee
et al. (2024) report limited interaction and critical reflection in Vision, while Omrani et al. (2025) find inadequate ICC integration in both Iran’s Vision 1 and Spain’s Network ESO 4; the former centers Islamic-Iranian culture, whereas the latter includes international content but often reproduces stereotypes. Rashidi and Najafi (2010) examined advanced-level ILI reading texts and found that only 7% concerned Islamic culture, even though the corpus included several other cultural categories. That figure is not directly comparable with the present elementary NE corpus: the earlier study analyzed advanced reading passages, whereas the present study codes cultural episodes across texts, visuals, and tasks. Differences in proficiency band and unit of analysis may therefore account for divergent source-culture totals.

Taken together, the literature identifies a recurrent but non-uniform tension: national materials often foreground source culture, but international series more strongly foreground target or global contexts; neither orientation alone guarantees tasks that develop ICC. Evidence for ILI materials is difficult to generalize across levels, and prior studies rarely combine ICC-component analysis with cultural-orientation coding in the same proficiency-matched corpus. The studies reviewed here have not compared Prospect, ILI NE, and AEF at A1–A2 or traced level-by-level progression within each series. These unresolved issues motivate the following four research questions:

1. To what extent do Prospect 1–3, ILI NE 1–3, and AEF Starter/1 embed ICC components across tasks?
2. How are source, target, and international cultures distributed across tasks in each series?
3. Do the three series differ significantly on ICC coverage and cultural orientation?
4. Within each series, is there clear progression across levels in ICC coverage and orientation balance?

Methods

Research Design

This study used a descriptive comparative content-analysis design with quantitative coding and exploratory inferential tests to examine how intercultural communicative competence (ICC) is represented in eight elementary-level student books: Prospect 1-3, ILI New Elementary (NE) 1-3, and American English File (AEF) Starter and 1. Content analysis was chosen over approaches such as critical discourse analysis or ethnographic observation because the goal here is to compare patterns across a defined corpus of published materials. Its primary purpose was to describe the frequency, form, and cultural orientation of opportunities designed into the books, not to infer learner competence or classroom effects. Byram's (1997) ICC framework and Cortazzi and Jin's (1999) cultural typology supported two complementary comparisons: competence-related opportunities and cultural referents. Counts were normalized for differences in book length and unit structure.

Corpus

The corpus was purposively selected to represent three institutional publishing streams available to Iranian learners: state-mandated national materials (Prospect), an institute-produced local series (ILI NE), and an international commercial series (AEF). Selection was based on institutional role, intended beginner/elementary placement, and sufficient overlap in communicative syllabus and proficiency range for a bounded comparison. Prospect's curricular status is documented; however, no adoption survey was conducted for ILI NE or AEF, so their inclusion should not be read as evidence of national prevalence. Likewise,
A1-A2 was used as a working band based on the series' level descriptions and was not independently verified through learner testing or lexical and grammatical profiling. Comparability therefore concerns intended elementary use, not equivalence in learner age, pages, units, or task density.

The analyzed corpus comprised the complete Student's Book editions listed in Table 1 and the reference list: Prospect 1-3 for Grades 7-9, three consecutive ILI NE elementary levels, and AEF Starter and 1 as the international comparator. Bibliographic identity was checked against each copy's title and copyright pages (title, authorship, edition, publisher, and year), and pages from different editions were not combined. Workbooks, teacher's books, tests, audio/video materials, and digital extensions were outside the corpus. The eight books contained 1,123 numbered pages and 72 publisher-labeled units: Prospect, 320 pages/21 units; ILI NE, 500/27; and AEF, 303/24. These unequal denominators were retained and addressed through page- and unit-normalized reporting. Table 1 summarizes the textbooks analyzed:

Table 1. Descriptive Information on Analyzed Textbooks

AEF 2 was excluded because the sampling rule required complete volumes within the working A1-A2 band. Although parts of AEF 2 may overlap upper A2, the volume extends into B1; including it in full would introduce above-band material, whereas extracting only its A2 content would violate the full-book rule and create selective sampling not used for Prospect or ILI NE. Consequently, AEF contributes two observed levels while the local series contribute three. Each book was treated as a book-level case and an ordered point within its own series, and the AEF progression analysis was limited to a single Starter-to-Book 1 comparison.

Analytic Tools and Frameworks

To examine the representation of ICC across the selected ELT textbook series, this study applied a combined analytical framework based on Byram’s (1997) model of ICC and Cortazzi and Jin’s (1999) typology of cultural content. Byram’s framework conceptualizes ICC through five interrelated components: knowledge of social groups and their practices, skills of interpreting and relating cultural perspectives, skills of discovering new cultural meanings and engaging in communication, attitudes of openness and curiosity, and critical cultural awareness. These categories informed the coding of textbook content at both textual and task levels. Rather than treating these as labels to be applied impressionistically, the study operationalized each component with an explicit definition. The coding matrix attached to the appendix summarizes the operationalization used during analysis. To further assess the nature and balance of cultural content, the study employed Cortazzi and Jin’s (1999) tripartite model, which distinguishes between source culture, target culture and international culture. Because the target culture should not be treated as a single, undifferentiated whole, Kachru's (1985) three-circles model, which itself recognizes the inner circle as internally diverse
(the United Kingdom, the United States, Australia, etc.), was considered while coding and analyzing the data. Source and international cultures retained their single codes. Each coding record also distinguished mode (textual, visual, task-based, or multimodal), explicitness (named cultural information versus a referent inferred from contextual cues), learner engagement (passive exposure/recognition versus active response, comparison, or interaction), and depth (surface products and practices versus values, assumptions, perspectives, or pragmatic norms).

The two frameworks are compatible because they operate at different levels of abstraction: Byram's model is competence-oriented (it describes capacities of the intercultural speaker that a textbook can stage opportunities for), whereas Cortazzi and Jin's typology is content-oriented (it classifies the referents on the page). Both rest on the communicative view of culture in language education that underpins the "intercultural speaker" ideal (Byram, 1997; Byram & Wagner, 2018) and both have been used in prior textbook research pairing cultural-origin coding with engagement or depth coding (Bose & Gao, 2022; Morady Moghaddam & Tirnaz, 2023).

Data Collection and Analysis Procedure

Each Student's Book was examined from its first through final publisher-labeled unit, including dialogues, readings, vocabulary and grammar sections, visuals, and communicative tasks. Printed pagination and labeled units supplied the denominators. The primary coding unit was a cultural episode: a bounded text, visual, learner prompt, or integrated sequence organized around one cultural referent or communicative purpose. A text and its accompanying image or task were counted as one multimodal episode when they functioned together; distinct referents or learner actions were recorded separately. Purely decorative images, incidental foreign names in decontextualized drills, and repetitions that added neither cultural information nor a new learner demand were excluded. A task was a bounded prompt requiring an observable learner response. Because all pedagogical tasks were not enumerated, an ICC-per-task denominator could not be calculated.

Coding proceeded in two linked passes. First, all eligible episodes were logged by book/unit/page, mode, explicitness, engagement, depth, and cultural orientation for RQ2. Second, episodes that provided an ICC opportunity were assigned one principal Byram component for RQ1. One primary coder conducted the full-corpus coding. During the pilot, selected units representing the three series were reviewed by a second applied-linguistics researcher familiar with ICC and ELT textbook analysis. Borderline cases were discussed until a single decision rule was recorded, and the codebook was revised before full coding. The second reviewer did not independently recode a predefined subsample; accordingly, Cohen's kappa or a comparable intercoder coefficient was not calculated. The pilot therefore strengthened codebook consistency but does not constitute statistical evidence of intercoder reliability.

Raw counts and within-book percentages were used for RQ1 and RQ2. To compare unequal books for RQ3, ICC counts were also expressed per 100 pages and per instructional unit; the higher-order share combined interpreting/relating, discovery/interaction, and critical awareness. Exploratory chi-square tests compared ICC-component composition and cultural orientation across series; a sensitivity test collapsed ICC codes into lower- and higher-order groups. Cramer's V was reported as effect size, followed by Holm-adjusted pairwise comparisons. Book-level ICC density was compared with a Kruskal-Wallis test on page-normalized rates and a unit-normalized sensitivity analysis (two-sided alpha = .05). For RQ4, progression was assessed descriptively across ordered books using normalized ICC density, higher-order share, normalized Shannon evenness (J) across the three orientations, and the combined target-plus-international share. Progression meant a consistent directional pattern in these book-level indicators; it did not measure the sequencing, conceptual difficulty, cognitive demand, or scaffolding of individual tasks. No frequency was weighted by subjective importance, and the two-level AEF pattern was interpreted as tentative rather than as a full trajectory.

Findings

RQ1: To what extent do Prospect 1–3, NE 1–3, and AEF Starter/1 embed ICC components across tasks?

Table 2 reports the distribution by book and series, together with each series' total and the share accounted for by higher-order components (interpreting/relating, discovery/interaction, and critical awareness), which require learners to compare, enact, or evaluate rather than merely recognize.

Table 2. ICC Component Distribution (with Series Totals and Higher-Order Share

Note. Higher-order share = (interpreting/relating + discovery/interaction + critical awareness) ÷ total. Critical cultural awareness was coded when a unit asked learners to take a stance on, question, or critically reflect on a cultural perspective or its underlying values (as distinct from recognizing or describing it).

Across the three series, NE returned the largest number of coded instances (73), followed by AEF (55) and Prospect (41). These are raw counts, not density measures: the series differ in corpus size (Prospect and NE span three books, AEF two), and books differ in length and task density, so the totals should not be read as ICC per unit of text. Cultural knowledge was the most frequent component in every series, consistent with elementary-level materials that foreground factual content; attitudes appeared with similar regularity. The components that demand deeper engagement — interpreting/relating, discovery/interaction, and critical awareness — were present throughout but concentrated differently: descriptively, AEF devoted the largest share of its ICC content to these higher-order components (38.2%), followed by NE (34.2%) and Prospect (24.4%).

To test whether these descriptive differences were statistically reliable, two exploratory analyses were run on the counts in Table 2. First, a chi-square test of independence on the 3 (series) × 5 (component) contingency table tested whether the composition of ICC components differed across series (not whether any series contained more ICC overall). The result was not significant, χ²(8, N = 169) = 3.69, p = .884, Cramér's V = .10. Because expected counts for critical awareness were small, the test was treated as exploratory and a second, more robust comparison contrasted lower-order components (knowledge and attitudes) with higher-order components (interpreting/relating, discovery/interaction, and critical awareness); this difference was also not significant, χ²(2, N = 169) = 2.09, p = .352, Cramér's V = .11, and pairwise comparisons between series did not reach significance
(all p > .15). These results mean that, on the present counts, AEF and NE contain a descriptively larger share of higher-order ICC components than Prospect, but the pattern should be read as a tendency rather than a confirmed series-level difference, and the analysis is likely underpowered given the modest number of higher-order instances.

In order to show exactly what those instances look like, Table 3 therefore gives a representative example per component per series, drawn from the inspected materials.

Table 3. Representative ICC Instances by Component and Series

The patterns in Tables 2 and 3 are consistent with the curricular orientations of the three series. Prospect, as the state-mandated series designed around Iranian cultural referents, concentrates its ICC content in cultural knowledge drawn from learners' own context — family and food in Prospect 1, the Isfahan lesson in Prospect 2, and festivals presented chiefly for recall in Prospect 3 — with comparatively few instances that ask learners to interpret, enact, or critique. NE and AEF, by contrast, pair source-culture content with international referents and more consistently stage comparison and interaction: NE moves from source-referent naming toward cross-cultural comparison (the biographical readings and the "A Trip Around the World" unit), while AEF embeds culture in interactive tasks such as role-plays and opinion exchanges. The summary of the abovementioned report is depicted diagrammatically in Figure 1.

Figure 1. Distribution of ICC Components across Textbook Levels

RQ2: How are source (Iranian), target (Anglophone), and international cultures distributed across texts, visuals, and tasks in each series?

This section draws on Cortazzi and Jin’s (1999) framework to categorize cultural content into three domains: source culture, target culture, and international culture. Each textbook was reviewed for references to places, practices, social customs, names, holidays, dialogues, and cultural scenarios, and these were coded based on the origin of the cultural material. Based on the information depicted in Table 4, the Prospect series is consistently source-centered: 41 of 52 episodes (78.8%) are Iranian, while target and international episodes account for 17.3% and 3.8%, respectively. Source representation is not limited to isolated names. It is reinforced across text and image through Iranian school and family settings, clothing such as the chador and manteau, the Isfahan lesson, and festival practices. By contrast, most external content is brief or embedded in an Iranian frame, such as the country-nationality drills in Prospect 2, the German tourist in Tehran, and the account of Ahmed from Turkey in the festival lesson. The few Anglophone references are mainly UK-related labels (e.g., England/British) or mixed/unspecified cues rather than developed US and UK contexts.

ILI NE has the most mixed numerical profile, although target culture remains the largest category (52.9%), with source and international culture each accounting for 23.5%. The supplied scope-and-sequence pages support this broader orientation through topics such as Mr. Amiri, Marie Curie, Pablo Picasso, Culture Shock, Personal Space, and tasks involving food in one's country, festivals, geographical places, and comparison. These pages verify topic and task breadth but do not permit an independent count of culturally identifiable visuals or a full assessment of within-unit depth. Accordingly, claims about specific diner settings, British correspondents, or other unverified scenarios have been removed.

AEF is oriented chiefly toward target (59.5%) and international culture (35.1%), with only two source-culture episodes (5.4%). The target category is internally diverse: US contexts include New York, Vermont, Chicago, and America: the Good and the Bad, whereas UK contexts are developed through Arriving in London, the British-English comparison, and the recurring Practical English sequence. International content is likewise disaggregated rather than treated as a single residual category: examples span Asia, continental Europe, Latin America, North Africa, and transnational topics such as Five Continents in a Day.

Table 4. Cultural Orientation Distribution across Each Textbook: Coded Episodes and Within-Book Percentages

Table 4 reports coded cultural episodes rather than undifferentiated token mentions. The within-book percentages describe cultural breadth, not pedagogical intensity: a brief label and a sustained cultural sequence are therefore interpreted through the separate mode-and-depth evidence in Table 5.

Table 5. Modal, Contextual, and Depth Disaggregation of Cultural Representation

Figure 2 uses a common 0-100% scale and equal-length bars, avoiding the visual inflation produced by comparing raw totals on a truncated or uneven scale. The contrast between Prospect's 41 source episodes and AEF's two is substantial, but the counts alone cannot establish protectionism or nationalism; they are also consistent with different publishing purposes - local identity maintenance in a national curriculum and a portable, globally marketed course in AEF.

Figure 2. Within-Series Distribution of Cultural Orientation

RQ3: Do the three series differ significantly on ICC coverage and cultural orientation?

ICC coverage and cultural orientation were analyzed separately: the former represents competence-related opportunities, whereas the latter represents the composition of cultural referents. Their raw counts were therefore neither pooled nor plotted on a common scale.

ICC Coverage

Table 6 counts a total of 41 for Prospect, 73 for ILI NE, and 55 for AEF. After adjustment for book length, AEF has the highest page-normalized rate (18.2 instances per 100 pages), followed by ILI NE (14.6) and Prospect (12.8). Per instructional unit, however, the order is ILI NE (2.70), AEF (2.29), and Prospect (1.95). The change in ranking confirms that cumulative totals alone cannot establish stronger ICC coverage.

Table 6. ICC Coverage Normalized by Book Length and Instructional Units

Figure 3. Descriptive ICC Density by Series

Prospect rises from 9.1 instances per 100 pages in Book 1 to 13.5 in Book 2 and 14.7 in Book 3. ILI NE similarly ranges from 12.0 in NE 1 to 14.5 in NE 2 and 16.9 in NE 3, while AEF increases from 17.2 in Starter to 18.9 in Book 1. Although the average ordering favors AEF on a page basis, the book-level ranges are not fully separated: ILI NE 3 (16.9) approaches AEF Starter (17.2), and Prospect 3 (14.7) is close to ILI NE 2 (14.5). The series averages therefore summarize overlapping book-level distributions rather than three uniformly distinct bands of ICC density.

Consistent with this overlap, the exploratory Kruskal-Wallis test found no statistically significant difference in page-normalized ICC coverage across the three series, H(2) = 4.25,
p = .119. The unit-normalized sensitivity analysis led to the same conclusion, H(2) = 1.23,
p = .540. Thus, the observed differences in density may be reported descriptively, but they do not provide sufficient evidence of a series-level difference in ICC coverage. This nonsignificant result should not be interpreted as proof that the series are equivalent: both tests have limited power because they are based on only eight ordered volumes, with unequal numbers of books across series. Moreover, the rates measure the frequency of coded ICC opportunities, not their instructional depth or effectiveness.

Cultural Orientation

Cultural orientation shows a clearer contrast because it concerns the composition of the 140 coded cultural episodes rather than the density of ICC opportunities. In Prospect, 41 of 52 episodes (78.8%) are source culture, whereas 9 (17.3%) are target culture and only 2 (3.8%) are international. ILI NE presents a more mixed profile: target culture is the largest category with 27 of 51 episodes (52.9%), while source and international culture each contribute
12 episodes (23.5%). AEF is also target-oriented, with 22 of 37 episodes (59.5%), but assigns a larger share to international culture (13 episodes, 35.1%) and very little to Iranian source culture (2 episodes, 5.4%) (see Table 4). Thus, ILI NE and AEF share a target-led profile, even though ILI NE retains more local grounding and AEF allocates more of its remaining content to international contexts. Prospect is the distinct case: approximately four fifths of its episodes are source-based, and target and international cultures together account for only about one fifth.

The association between textbook series and cultural orientation was statistically significant, χ²(4, N = 140) = 58.31, p < .001, with Cramer's V = .46, indicating a sizeable relationship between the series and the type of culture represented. Holm-adjusted pairwise tests clarified the source of this omnibus result: Prospect differed from both ILI NE (p < .001) and AEF (p < .001), whereas the difference between ILI NE and AEF did not reach the
.05 level (p = .061). The significant overall finding is therefore attributable chiefly to Prospect's strongly source-centered distribution, rather than to a statistically established difference between the two target-led series. This test identifies a difference in proportional orientation only; it does not determine whether one orientation is pedagogically preferable or whether its episodes are deeper, more multimodal, or more interactional.

RQ4: Within each series, is there clear progression across levels in ICC coverage and orientation balance?

Table 7 reports the ICC counts, rates per 100 pages and per instructional unit, the higher-order ICC share, orientation evenness, and the proportion of target plus international episodes. Because Prospect and ILI NE contain only three observed levels and AEF only two, these patterns are descriptive rather than inferential trends.

 Prospect Series

The ICC count rises from 8 in Prospect 1 to 13 in Prospect 2 and 20 in Prospect 3. The same direction remains after controlling for book length: coverage increases from 9.1 to 13.5 to 14.7 instances per 100 pages and from 1.00 to 1.86 to 3.33 per unit. The higher-order share also rises from 12.5% to 30.0%. Thus, the results do not support a Book 3 dip, although the page-normalized increase becomes smaller after Book 2. Orientation changes more modestly. Source-culture episodes decline from 85.7% to 75.0%, the non-source share rises from 14.3% to 25.0%, and evenness increases from .37 to .63, mainly between Books 1 and 2.
Yet Prospect 3 remains three-quarters source-based, and international content never exceeds 5.6% (see Table 4); the sequence therefore shows increasing ICC coverage and some diversification, but not a fully balanced orientation.

ILI New Elementary Series

ILI NE shows the clearest three-level rise in normalized ICC coverage. Counts increase from 19 to 24 to 30; the corresponding rates rise from 12.0 to 14.5 to 16.9 per 100 pages and from 2.11 to 2.67 to 3.33 per unit. The higher-order share increases only moderately, from 31.6% to 36.7%. The cultural distribution, however, does not become more even. As source culture falls from 33.3% to 15.8%, target culture grows from 46.7% to 57.9% and international culture from 20.0% to 26.3%. Consequently, the non-source share rises from 66.7% to 84.2%, while evenness declines from .95 to .87 because target culture increasingly becomes the largest category. ILI NE therefore progresses in ICC coverage and broadens non-source exposure, but the latter shift should not be described as improved orientation balance.

 American English File Series

AEF shows an upward step from Starter to Book 1: ICC counts rise from 23 to 32, page-normalized coverage from 17.2 to 18.9, unit-normalized coverage from 1.92 to 2.67, and the higher-order share from 34.8% to 40.6%. With only two observations, however, this is preliminary evidence of change, not a strong developmental trajectory. Cultural orientation is essentially stable: source culture changes from 5.9% to 5.0%, target culture from 58.8% to 60.0%, and international culture from 35.3% to 35.0%. Evenness decreases slightly from
.77 to .75 while the non-source share remains near 95%. The two books therefore retain a target/international-led profile rather than moving toward a more even distribution that includes learners’ source culture.

Table 7. Level-by-Level Indicators

Note. Higher-order ICC combines interpreting/relating, discovery/interaction, and critical awareness. J is normalized Shannon evenness across source, target, and international episodes; higher values indicate a more even three-way distribution. Non-source combines target and international episodes.

Table 8. Summary of Within-Series Progression

Taken together, Prospect and ILI NE show consistent three-level increases in normalized ICC coverage, whereas AEF shows a single two-level increase. The orientation results are not parallel: Prospect becomes somewhat more even but remains strongly source-dominant, ILI NE becomes less even as target culture expands, and AEF remains stable and source-poor. This conclusion is limited to book-level content indicators. Normalizing by pages and instructional units reduces the effect of unequal book length, but it cannot show whether ICC is clustered in a few units or sequenced arbitrarily.

Discussion

The findings are best interpreted as three profiles within this bounded corpus, not a hierarchy of textbook quality or a pattern representative of Iran's entire elementary ELT system. The clearest contrast concerns cultural orientation: Prospect is strongly source-centered, whereas ILI NE and AEF are target-led. AEF and ILI NE have descriptively larger higher-order shares than Prospect, but the evidence does not establish more effective ICC learning.

In Prospect, familiar Iranian settings can make elementary English accessible and help learners describe their own lives, especially in a national curriculum shaped by religious and cultural sensitivities. The limitation is not Iranian content itself, but its weak conversion into relational work: lessons on Isfahan or festivals provide knowledge without consistently asking learners to anticipate another perspective, explain a practice, or compare interpretations. This resembles Morady Moghaddam and Tirnaz's (2023) contrast between Vision and the more internationally oriented Top Notch and accords with reports of limited ICC support in Iranian state textbooks (Banaruee et al., 2023; Rezaee et al., 2024). Prospect's normalized ICC coverage and higher-order share rise across all three books. After Book 2, the page-based gain becomes smaller and cultural evenness nearly plateaus: Prospect 3 adds ICC opportunities but diversifies orientations less readily.

ILI NE retains Iranian referents while extending target and international exposure and shows the clearest three-level increase in normalized ICC coverage. Yet variety is not the same as balance: target-culture representation grows as source culture declines, so orientation evenness falls from NE 1 to NE 3. Its higher-order share rises only modestly. Culture Shock, Personal Space, and A Trip Around the World create openings for comparison, but topic presence does not ensure sustained pragmatic learning. A task remains surface-level when learners need not choose among refusals, clarifications, apologies, politeness strategies, or repair moves and consider their effects. The findings thus partly support Hosseinzadeh et al. (2022), who reported broader cultural content in international and localized materials, while separating breadth from task depth. They are not directly comparable with Rashidi and Najafi's (2010) low figure for Islamic content because that study examined advanced reading passages rather than multimodal elementary episodes.

AEF has the largest descriptive higher-order share and page-normalized rate, although neither difference was statistically established. Its pragmatic work appears in checking into a hotel, service role-plays, and opinion exchanges about America: The Good and the Bad. These activities give beginners scripts for greetings, requests, clarification, turn-taking, and context-sensitive responses rather than cultural facts alone. AEF’s source-poor profile need not alienate learners: a "window on the world" may match their expectations or motivate them; Köksal and Ulum (2021), for example, found a strong preference for target-culture content among Turkish learners. The concern is not that AEF must reproduce a fixed proportion of Iranian content, but that it seldom asks learners to use their own repertoires for comparison. This echoes global coursebooks' selective representation (Keles & Yazan, 2020) without supporting the stronger claim that local identity is erased.

Across the series, the more defensible concern is the depth and sequencing of engagement. The findings parallel Abid and Moalla's (2020, 2022) observation that cultural contact and variety do not necessarily develop the intercultural speaker and resemble Yook and Kim's (2024) evidence of limited textbook support for ICC despite curricular aspirations. That study documents uneven and limited evidence, not deliberate avoidance, of deeper intercultural work. Moreover, none of the present series progresses in both normalized ICC coverage and orientation balance: Prospect diversifies modestly, ILI NE becomes increasingly target-led, and AEF shows only one upward step while its orientation remains stable. This variation cautions against treating a series as uniform or higher counts as systematic scaffolding. In Prospect, teachers can extend Isfahan, nationality, and festival material with "What might a visitor misunderstand?", then require clarification or repair in the tourist-receptionist role-play. In ILI NE, Culture Shock, Personal Space, and celebration tasks can introduce alternative refusals, apologies, or disagreement strategies for comparison and role-play. In AEF, teachers can add an Iranian equivalent to global scenarios, compare expectations, and practice explaining a local custom to a non-Iranian interlocutor.

Byram's (1997) categories distinguish recognition from interpretation, interaction, and evaluation, but can make culture appear more stable and nationally bounded than learners' identities and encounters (Dervin, 2016; Risager, 2007). Cortazzi and Jin's (1999) categories are likewise coding heuristics, not discrete cultural realities. Xu et al.'s (2025) Hybrid ICC Model highlights what this print-based analysis cannot capture: digital literacy, multilingual identity, cultural adaptability, and power relations. Even an AEF task coded as higher-order may rehearse a useful routine while leaving unexamined whose English or norms are treated as standard. Conversely, a low count cannot show that learners fail to develop ICC through teachers, peers, media, or other experiences.

This need for mediation is not proof that teachers are simply left on their own. Tavakoli (2021) showed that textbooks contribute strongly to Iranian institute teachers' subject and didactic competence and professional identity; uneven ICC prompts therefore increase the professional judgement needed to turn content into learning. The practical response is institutional support: concise teacher-guide notes, adaptable task banks, model language for sensitive comparisons, and training in reflection without stereotyping.

Conclusion

This study compares eight A1-A2 student books from three series and distinguishes cultural orientation from ICC opportunity. Cultural orientation varies more clearly than the amount or composition of ICC opportunity, and the significant difference is driven chiefly by the source-centered Prospect series, whereas differences in normalized ICC coverage and component composition are descriptive rather than statistically established. Moreover, none of the series progresses in both ICC coverage and orientation balance. Prospect provides strong local anchoring but limited non-source breadth and fewer higher-order opportunities; ILI NE offers broader exposure but becomes increasingly target-led; and AEF descriptively provides more pragmatic and higher-order work while leaving little space for learners to articulate Iranian experience. The issue is not a choice between local identity and wider participation, but whether representation is connected to supported comparison, interaction, and reflection.

The implications remain at the level of materials design rather than claims about classroom effects. Because Prospect also serves basic literacy, grammar, and national curricular purposes, revision need not displace these aims. A recurring micro-sequence could preserve these aims while adding one age-appropriate prompt per unit: notice a difference, explain an Iranian practice in simple English, compare interpretations, and rehearse clarification or repair. ILI NE units on celebrations, personal space, or culture shock could pair alternative refusals, apologies, and disagreement strategies with guided role-play. AEF users and materials adaptors could add an Iranian counterpart to global scenarios and ask learners to explain local expectations to a visitor. Teacher guides could support such work through ready-to-use prompts, model language, and options for culturally sensitive contexts. These are design proposals suggested by gaps in the books; whether teachers can enact them and whether learners benefit remain empirical questions.

These conclusions are bounded by a purposively selected, single-country corpus at one proficiency band. The analysis examines textbooks rather than classroom mediation or learner outcomes, includes only two AEF levels, and does not measure task complexity, cognitive demand, within-unit sequencing, or scaffolding. Page- and unit-normalization reduces differences in book length but cannot provide an ICC-per-task rate, and coding subjectivity cannot be fully eliminated without an independent reliability coefficient. The findings therefore characterize these eight books, not the whole Iranian ELT system or a universally applicable curriculum. Future research should combine broader textbook samples with classroom observation, teacher and learner perspectives, and performance-based measures. It should also test short, scaffolded revisions to determine whether they improve pragmatic choice, cultural interpretation, and reflective comparison at beginner levels. Within these limits, the study provides a country- and level-specific basis for connecting cultural representation with feasible opportunities to interpret, interact, and reflect.

Acknowledgements

We thank all EFL professors and students who did not withhold their valuable guidance from us.

Declaration of Conflicting Interests

The authors report that there are no competing interests to declare.

Funding

Not applicable

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, upon reasonable request.

Appendix 1

Coding matrix. Operationalization of Byram's (1997) ICC Components