Despite their intricate relationship, cohesion and coherence are fundamental elements for successful textual communication. According to Halliday and Hasan (1976) , cohesion indicates an inner semantic chain that works behind the scenes and facilitates coherence. It refers to the interconnections that relate sentences in a particular text. Major grammatical and lexical devices including reference, ellipsis, substitution, conjunctions, reiteration and collocations are what enables the writer to establish the internal relationships among sentences and provide information flow and dramatic unity. Coherence however, indicates the high quality of a text sentence by sentence. According to Todirascu et al. (2013) and Alsaawi (2016) , no matter where texts are from, the ones with high consistency level would be easier to understand. Likewise, McCulley (1985) and Ramadan (2003) contributed to the idea that cohesion ultimately leads to coherence, the key feature of a good style of writing. Brown and Yule (1983) also point out the importance of coherence which by itself may not ensure cohesion. Cohesive ties can exist beyond a text in the absence of the deeper semantic connections required to produce true coherence. Carrell (1982) and Brown and Yule (1983) recognize that the connections go beyond only the linguistic markers and are of a deeper semantic level. Celce-Murcia and Olshtain (2000) explain that coherence is what contributes to the unity of various sentences and paragraphs, that in turn ensures local and global unity of the text. O’Rourke, Calvo and McNamara (2011) state that a good text (or an ideal one) uses a combination of cohesive devices which lead to the production of a fluent discourse, coherent with sound logical judgments across longer texts.

The CEFR and Bloom’s Taxonomy serve as valuable tools in EFL writing instruction although they address distinct aspects of learning ( Anderson et al., 2001 ; Council of Europe, 2020 ). Understanding their relationship can empower teachers to design more effective learning experiences that target both language proficiency and deeper cognitive development. The CEFR provides a comprehensive framework for describing language proficiency across Europe and increasingly, on a global scale ( Council of Europe, 2020 ). It outlines six proficiency levels (A1 - C2) with detailed descriptors for reading, writing, listening, and speaking skills. These descriptors specify what learners can understand and do at each stage. In the context of writing, the CEFR helps teachers set learning objectives aligned with specific skill sets at each level ( Alderson et al., 2015 ). Additionally, the CEFR offers a foundation for assessment and evaluation by providing benchmarks based on “can-do” statements outlining learner capabilities ( Dylan & Marnie, 2017 ). Bloom’s Taxonomy, on the other hand, focuses on the cognitive processes involved in learning ( Anderson et al., 2001 ). It categorizes learning objectives into six levels of increasing complexity: remembering, understanding, applying, analyzing, evaluating, and creating. This framework empowers teachers to design writing tasks that require learners to engage in different levels of thinking ( Ferris et al., 2017 ). By incorporating Bloom’s Taxonomy, teachers can promote the development of critical thinking and higher-order thinking skills crucial for effective writing ( Littlewood, 2012 ).

The CEFR and Bloom’s Taxonomy can be used cooperatively to create a more comprehensive approach to EFL writing instruction. First, with respect to mapping CEFR Levels to Bloom’s Taxonomy, while not a perfect one-to-one correspondence, a rough mapping can be established. Lower CEFR levels (A1 - A2) primarily focus on remembering and understanding basic vocabulary, grammar and information in familiar contexts, while higher levels (B1 - C2) gradually incorporate more complex skills like applying knowledge in new situations, analyzing texts, and evaluating arguments ( Council of Europe, 2020 ). Nevertheless, creating, the highest level in Bloom’s Taxonomy, does not correlate to any specific CEFR level. It can be present at various levels depending on the complexity of the creative language task and the language proficiency ( Council of Europe, 2020 ). Second, when designing tasks with depth, using the CEFR descriptors for a specific level, teachers can craft writing tasks that incorporate different cognitive demands from Bloom’s Taxonomy. For example, a B1 writing task could require learners to understand (summarize) a news article while also applying their knowledge by writing a simple email based on the information presented ( Ferris et al., 2017 ).

When combined in EFL instruction, the CEFR and Bloom’s Taxonomy offer diverse benefits. Combining these frameworks allows for instruction that targets both the linguistic skills (CEFR) and the cognitive processes (Bloom’s Taxonomy) required for effective writing. This ensures learners develop not just proficiency in using the language but also the ability to think critically and express themselves effectively at a deeper level ( Littlewood, 2012 ). Moreover, by addressing both ‘what’ (language skills) and ‘how’ (thinking processes), this combined approach fosters a deeper understanding and a more meaningful learning experience for EFL learners. They not only acquire language proficiency but also develop the critical thinking skills necessary to navigate various writing contexts. Besides, by incorporating Bloom’s Taxonomy, teachers can go beyond simply teaching grammar and vocabulary. They can encourage learners to analyze information, evaluate arguments, and create their own unique perspectives, leading to a more sophisticated approach to writing ( Alderson et al., 2015 ).

The researcher adopts the mixed-method approach to collect quantitative and qualitative data, using three comparison-contrast essays per student in each level. This helps provide insight into the students’ performance and determine the improvements in coherence as well as lexical (reiteration and collocation) and grammatical (reference, ellipsis, conjunction and substitution) cohesion.

The subjects of the study were three groups of twenty-five students enrolled at the various faculties at Beirut Arab University. All students have similar characteristics as they come from various educational backgrounds in Lebanon and the region, their mother tongue is Arabic and their first foreign language is either English or French. All were newly admitted to Beirut Arab University and assigned the Intensive English language courses INTE 101, INTE 103 and INTE 104 during fall 2023-2024. The choice of these groups was not random; students were intentionally selected from these courses since as they align with the Bloom’s three levels: Lower, Middle and Higher. All participants agreed to be part of the study and were ensured that their identities shall remain confidential. They were also given the freedom to withdraw from the study at any phase.

The instruments used to collect data included three comparison-contrast essays per student. These essays were evaluated against a rubric (Appendix A) developed by the researcher. This rubric has six categories: Content, Coherence 1 (organisation of the essay), Coherence 2 (connectedness between ideas, sentences and paragraphs), Grammatical Cohesion, Lexical Cohesion and Language. It is used to examine the students’ performance in the comparison-contrast essay. The reliability and validity of the rubric were checked and confirmed (Appendix A-1). During the study, students used the CCCC writing guide as a reference for analysing and constructing their comparison-contrast essays. This guide (Appendix B) is designed by the researcher to help students develop a comparison-contrast essay with strong cohesion and coherence. The validity and reliability of the CCCC guide were examined and confirmed through piloting (Appendix B-1).

Three textbooks were used to present to students the structure and organisation of the comparison-contrast essay as well as the various cohesive devices of concern to this study: Effective Academic Writing, The Researched Essay Book 3, English Grammar in Use (CEFR B1 - B2) and English Collocations in Use (CEFR B1 - B2).

The study plan comprised three stages. The Presentation-Practice-Production (PPP) process was employed in class. The first stage (Present) started with writing the first essay (E1) to check students’ background knowledge and performance before explicit instruction. This stage lasted for three weeks during which the comparison-contrast essay writing process, coherence, lexical and grammatical cohesion and the CCCC writing guide were introduced. In stage two (Practice & Produce), the teacher exposed the students to authentic compare-contrast essays which they examined and analyzed using the writing guide. Students were also given focused practice on the skills of coherence and cohesion. This stage lasted for three weeks and ended with the guided writing activity essay 2 (E2). In stage three (Produce), students produced essay 3 (E3) on their own, using the CCCC guide and consciously applying the rules of cohesion and coherence they had already learned. Essay topics were adopted form the CCCC guide.

Each student in the three courses was scored on the competencies in the scoring rubric [content; Coh1 (coherence1/organisation of the essay); Coh2 (coherence2/ connectedness between ideas, sentences and paragraphs); CohG (grammatical cohesion); CohL (lexical cohesion); Lang (language) and Total (overall score)] across the three essays E1, E2, and E3. Data collected from the three essays were entered and analysed using Statistical Package for Social Science software (SPSS17).

For INTE101, statistics revealed that the total and competency mean scores increased across the three essays ( Table 2 ).

The score difference between E1, E2, and E3 were significant in every competency as evident by all p-values falling below the significance level of 0.05: For example, the results of Lang scores were as follows: F (2, 48) = 35.649, p ≤ 0.001, partial η² = 0.598 ( Table 3 ).

Post hoc analyses using the Bonferroni post hoc criterion for significance indicated that the average scores for all except one of the competencies increased from E1 to E2 to E3. For example, Coh1 score in E2 was significantly higher than in E1 (Difference = 0.520, p-value ≤ 0.001), and significantly higher in E3 than in E2 (Difference = 0.320, p-value ≤ 0.001). However, CohL score showed no significant increase between E2 and E3 (Difference = 0.160, p-value = 0.129). All in all, the results indicate that the implementation of the CCCC guide had a significant positive effect on individual student performance in almost every competency.

Statistics showed that the total and individual competency mean scores consistently increased across all three essays ( Table 4 ).

The score differences between E1, E2, and E3 were significant in each of the competencies as evident by all p-values falling below the significance level of 0.05. For example, the results of the Total scores were as follows: F (2, 48) = 146.522, p ≤ 0.001, partial η² = 0.859 ( Table 5 ).

Post hoc analyses using the Bonferroni post hoc criterion for significance indicated that the average score of CohG and the total score increased across the three essays. For instance, the CohG score in E2 was significantly higher than in E1

(Difference = 0.320, p-value = 0.008) and in E3 higher than in E2 (Difference = 0.640, p-value = 0.001). However, while the average scores for the remaining competencies (Coh1, Coh2, CohL, Lang, and Content) did not significantly increase from E1 to E2, they still showed an overall increase in E3. For example, while the average Coh1 score did not significantly differ from E1 to E2 (Difference = 0.120, p-value = 0.249), it increased significantly from E2 to E3 (Difference = 0.480, p-value ≤ 0.001). These results indicate that implementation of the CCCC guide in INTE103 had a positive effect on student performance.

Descriptive statistics revealed an increase in the scores of all competencies from E1 to E2 and further from E2 to E3 ( Table 6 ).

The score differences between E1, E2, and E3 were significant in the competencies of Coh1, Coh2, CohG, CohL, and Total as evident by p-values falling below the significance level of .05: For example, the results of the Total scores were as follows: F (2, 48) = 50.174, p ≤ 0.001, partial η² = 0.676 ( Table 7 ).

Post hoc analyses using the Bonferroni post hoc criterion for significance indicated that, for competencies that showed a significant change in score across the three essays, the score difference between the first and second or the second and third essays wasn’t always significant, while the overall difference between the first and third essays was always significant: For example, Coh1 mean score difference between E1 and E2 was 0.2, with a p-value of .288, but the mean difference between essays 1 and 3 was 0.6, with a p-value of 0.002. On the other hand, for competencies that showed no significant results (Lang and Content), no significant difference was found between any pair of essays. All in all, the results indicate that the implementation of the CCCC guide in INTE104 positively affected individual students’ overall and competency performance except for Content and Lang.

Assessment Rubric for Comparison-Contrast Essay

1 = very poor 2 = Poor 3 = Fair 4 = Good 5 = Very Good

Overall Mark: ______/30

Developing an effective rubric is crucial for fair and accurate assessment of student writing. Here’s a breakdown of key validation strategies to ensure the rubric is reliable, valid, clear, unbiased, and user-friendly:

1) Reliability Testing:

Inter-rater Reliability: Two INTE 103 instructors were randomly selected to assess five writing samples for INTE 103 students using the rubric. The level of agreement between both raters was calculated to assess consistency using percent agreement as shown in Table A1 . An 80% agreement means that the majority of the students received the same scores by the both assessors.

Intra-rater Reliability: One of the assessors was randomly selected to score the same writing samples two times (T1, T2) at two different intervals. The correlation coefficient was calculated to check the consistency of scores in order to ensure the rubric yields consistent results over time. A correlation of r = 0.9 suggests a strong, positive association between two variables ( Table A2 ), which means that the scores are very close to each other at both intervals of time.

2) Validity Testing:

Content Validity: The rubric was accurately reviewed and evaluated by the advisor and subject-matter experts—in this case three INTE 103 instructors—to examine if it reflects the intended learning objectives and writing skills to be assessed (herein comparison, contrast, coherence, cohesion).

Construct Validity: Students’ performance on the essay using the rubric was analyzed and compared with another measure of the same essay (using a checklist) to see if the rubric captures the targeted construct. Therefore, a t-test was conducted to determine if there is a significant difference between the mean scores of two groups of essays and how they are related. A p-value of 0.368565057 (greater than 0.05) indicates that the result is insignificant. This means that the rubric yielded similar results to the other measure ( Table A3 ).

3) Clarity and Usability:

Peer Assessment: Five INTE 103 students and an instructor were randomly selected to assess the understandability of the rubric’s language and criteria. The essays were graded by the instructor on the one hand and by the students as well. This helped check if students were able to understand the wording and the structure of the rubric. However, all peer assessment scores were higher than those of the instructor ( Table A4 ). A p-value of 0.004636 (smaller than 0.05) indicates that the result is significant. This means that the rubric did not yield similar results by the instructor and the students.

Interviews: Based on the results of the t-test, the researcher interviewed both the students and the instructor to identify any areas of confusion or difficulty in interpreting the criteria and descriptors. The instructor found the rubric clear while the students found difficulty interpreting the sections on cohesion and coherence. Therefore, the wording and structure of these sections in the rubric were refined for clarity based on the feedback received.

4) Bias and Fairness:

Review for Bias: The advisor analyzed the rubric for language or criteria that would favor or disadvantage certain student groups, including cultural bias or focusing on specific writing styles.

Multiple Perspectives: The perspectives of two academic experts (INTE 103 instructors) were included when developing and reviewing the rubric to minimize bias.

5) User Feedback:

Student Survey: After updating the rubric based on the students’, instructors’, raters’, experts’ and the advisor’s recommendations, a student survey was conducted. The five students involved in the peer-assessment activity were asked to reply to a 7-likert scale survey concerning the usage of the rubric in assessing the comparison-contrast essays. The survey includes five statements about the rubric’s clarity, connection to the writing task, usefulness, fairness, and overall experience.

Tables A5-A9 depict the results of the student survey concerning the implementation of the rubric in peer assessment. All results indicate that the students either agreed or strongly agreed to each statement with the percentages of 40 and 60 or vice versa.

Educator Interviews: An interview was conducted with the instructor to understand her experience using the rubric. The following questions were discussed:

General Use:

Fairness and Assessment:

Additional Feedback:

A) Comprehensive Documentation

Purpose: The CCCC guide aims at helping students write a comparison-contrast essay with strong cohesion and coherence.

Intended Use: The guide was used as one of the instruments in this study to check its efficiency in enhancing the INTE 101, 103 and 104 students’ comparison-contrast essays.

Structure: The CCCC guide integrates comparison-contrast writing with cohesion and coherence in alignment with Bloom’s Taxonomy and CEFR levels. It is divided into three levels, lower, middle and upper. Each part includes ersonalized pedagogical strategies that match with its level of difficulty. These strategies include task examples, cohesion skills, coherence skills as well as recommended teaching-learning approaches.

B) Examples and Explanations

Bloom’s Lower Levels:

Key Considerations:

Bloom’s Middle Levels

Key Considerations:

Cohesion/Coherence:

Bloom’s Higher Levels

Key Considerations:

Cohesion/Coherence:

A) Purpose of Validation: The validation process is essential to ensure that the CCCC Writing Guide is an effective tool for enhancing the quality of writing, specifically in terms of comparison, contrast, coherence, and cohesion. This involves confirming that the guide is both reliable and valid.

B) Validation Process

1) Writing Samples Comparison: The pre-test and post-test writing samples were analyzed using the rubric to identify improvements in areas targeted by the guide.

2) Results of the Pretest: Before implementing the CCCC guide, the students had been tested on the relevant competencies through a comparison-contrast essay, resulting in a pre-test score for each. Descriptive statistics were obtained to determine students’ performance in each competency and in the essay as a whole.

In Coh1, 3 students (50%) received 1/5 on the pre-test essay, and 3 (50%) received 2/5. This means that all scores were below average (3/5). Results are displayed in Table A10 .

In Coh2, four students scored 1/5 (66.66%) and two students scored 2/5 (33.33%). This means that none of the students received an average score on Coh2 in the pretest essay. Results are shown in Table A11 .

Table A12 presents the results of CohG. Two students received 1/5 (33.33%), three students received 2/5 (50%) and one student got 3/5 (16.66%). This indicates that the majority of the students performed poorly on CohG in the pretest.

As for CohL, three students received 1/5 (50%) while the other three students got 2/5 (50%). This reveals that all students could not achieve average performance on CohL in the pretest ( Table A13 ).

The distribution of the students’ total scores in the pre-test essay is depicted in Table A14 . One student received 9/30 (16.66%), one student got 11/30 (16.66%) and three students got 12/30 (50%). This means that all students failed the pretest essay.

1) Results of the Post-test: After implementing the CCCC guide, the students were tested on the relevant competencies through a comparison-contrast essay similar to the pre-test, resulting in a post-test score for each. Descriptive statistics were obtained to determine students’ performance in each competency and in the essay as a whole.

In Coh1, two students scored 1/5 (33.33%) and four students scored 3/5 (66.66%) in the post-test. This means that more than half of the students achieved the average score on Coh1 in the post-test as shown in Table A15 .

In Coh2, one student (16.66%) received 2/5 while the other five students (83.33%) received 3/5. This indicates that the majority of the students received a passing grade on Coh2 in the post-test essay. Results are displayed in Table A16 .

Table A17 presents students’ scores on CohG in the post-test essay. Two students (33.33%) got 2/5, three students (50%) received 3/5 and one student (16.66%) received 4/5. This indicates that the majority of the students performed well on CohG in the post-test essay.

The results of CohL are similar to those of CohG as displayed in Table A18 . Most of the students did well on CohL in the post-test essay.

The distribution of the students’ total scores in the post-test essay is described in Table A19 . One student received 14/30 (16.66%), one student got 16/30 (16.66%), another student got 18/30 (16.66%), two students received 19/30

(33.33%) and one student scored 20/30 (16,66%). This means that 66.66 students passed the post-test while the scores of the remaining students showed significant improvement.

Table A20 presents the students’ score progress across the pre-test to the post-test essay as represented in the score mean and standard deviation. For instance, the students’ mean total scores in the post-test increased by 5.8 points in comparison with their mean total scores in the pre-test.

1) Pilot Group Student Survey: The pilot group students were required to reply to a 7-likert scale survey concerning the implementation of the CCCC Guide in writing a comparison-contrast essay. The survey includes five statements about the guide’s clarity, organization, and helpfulness:

Survey Analysis

The analysis of students’ replies to the first statement S1 in the survey reveals that one student (16.66%) somehow agreed, three students (50%) agreed and two students (33.33%) strongly agreed that it was easy to understand the information in the guide ( Table A21 ).

As for the second statement S2, one student (16.66%) somehow agreed, three students (50%) agreed while two students (33.33%) strongly agreed that the instructions were clear and easy to follow. Results are shown in Table A22 .

Concerning statement three S3 ( Table A23 ), one student (16.66%) somehow agreed, two students (33.33%) agreed and three students (50%) strongly agreed that the guide provided enough examples to illustrate the concepts.

Students’ replies to statement four S4 varied greatly ( Table A24 ). One student (16.66%) disagreed, another (16.66%) somehow disagreed, another student (16.66%) neither agreed nor disagreed, one student (16.66%) somehow agreed

while two students (33.33%) agreed that the guide contains no sections that are confusing or difficult to understand.

Table A25 depicts the results of the students’ replies to the fifth statement S5. Two students (33.33%) agreed while the other four students (66.66) strongly agreed that the guide was useful in helping them improve their writing.

1) Educator Feedback: A One-on-one interview with the teacher was held to discuss her experience using the guide in the classroom. The following questions were discussed:

Revisions Based on Data: The collected feedback was analysed to identify common themes and areas for improvement. The pre- and post-test results confirm that the CCCC guide has generally helped students develop their comparison-contrast essay writing. However, based on the students’ answers in the survey, it appeared that they were confused about some parts of the guide as is clear in statement 4 of the survey: The guide contains no sections that are confusing or difficult to understand. Hence, the researcher asked the students to identify the sections they found difficult. Their replies matched with the instructor’s. All of them recommended that more examples be added especially under task examples, coherence and cohesion. Moreover, they required that the section on pedagogical strategies be clarified with more explanation and examples. Based on the pilot test results, the researcher revised the writing guide to improve its clarity, usability, and effectiveness in achieving its intended outcomes. All recommendations were considered and the necessary modifications were done.