The Effect of Reciprocal Teaching on Third Year Nigerian ESL Students’ Mastery of William Faulkner’s “A Rose for Emily”

This true experimental study examined the effects of reciprocal teaching on 3 year Nigerian university students’ mastery of “A Rose for Emily,” an English short story famously known for its intriguing plot but difficult narrative style. The subjects were a class of 60 students who were randomly assigned to either the treatment (n = 30) or control group (n = 30) based on the matching of their pretest scores. The experiment was conducted over 12 weeks where the treatment group studied the short story using reciprocal teaching that comprised the activities of summarizing, questioning, clarifying and predicting, while the control group was taught the same story using direct instruction. The subjects' mastery of the literary text was measured in terms of vocabulary, comprehension and recall using a self-developed test. The reliability of the measures ranged from α = 0.60 to α = 0.80. Descriptive statistics and independent samples t-tests were used to analyse the data. The results indicated that RT was very effective in enhancing the subjects’ recall and comprehension of the story and their acquisition of new vocabulary. The treatment group exhibited a significant increase of 67.6 points in their mastery of the short story compared to a significantly lower gain of 44.9 points by the control group. The effect size of the treatment was very large at Cohen’s d = 4.76. The results supported the use of active text processing strategies and social group interaction in facilitating students’ learning of literature as purported by Vygotsky’s social constructivist theory.


INTRODUCTION
Students learning English as a second language (ESL) frequently encounter difficulties in reading and comprehending English Literature, especially that from the Old English (c. 450 -1066), Renaissance (1500Renaissance ( -1660 and Romantic (1798Romantic ( -1837 periods. Literature from these periods contains vocabulary that is often archaic and sentence structures that can be quite complex. Due to these factors and also to the subject matter they deal with, reading and understanding literary works can be daunting. This article demonstrates the potential role of reciprocal teaching, a collaborative reading activity performed by students in small groups, in alleviating ESL learners' difficulty in dealing with English literature. The article begins with a discussion of the challenging short story, "A Rose for Emily," to illustrate why the reading of the story is potentially difficult for ESL learners. It then proceeds to explain the nature of reciprocal teaching and how the strategy works to alleviate difficulties with reading comprehension.

Reading "A Rose for Emily" as a Literature Piece
William Faulkner, an early 20th century American writer, wrote and published the famous short story, "A Rose for Emily," in April 1930. The tale was set in a fictional town called Jefferson, Mississippi and told the story of a white aristocratic woman's gradual descent into insanity after experiencing a continuous series of tragic events in her life. Miss Emily Grierson, the story's female protagonist, was a rich and beautiful young woman who had to endure her father's suppressive and dominant control of her life, and due to this oppressive control, she was deprived of courtship and male companionship throughout her youth. As a result, by the age of 30, she was still unmarried, something that society greatly frowned upon during that era. At the same time, her wealth and aristocracy led to a life of estrangement and isolation from the rest of her community. Her father's sudden death caused Miss Emily to descend into a long period of shock and depression. She refused to accept that he had died and kept his dead body in the house for three days. Her strange behavior caused the townspeople to suspect that she had lost her mind and gone insane.
Following her father's passing, Miss Emily shut herself out from the rest of the world, only to reappear in the midst of society with a new romantic interest, Homer Barron, after months of hiding away. Naturally, the townspeople expected the two to marry soon as prolonged courtship was not an acceptable norm of the time. However, not long after being seen several times with Miss Emily, Homer Barron mysteriously disappeared and was never seen again by any of the town's folks. It was assumed that he had secretly and silently left Jefferson for good. About three decades later, after Miss Emily's demise at the age of 60, the townspeople discovered the skeletal remains of a man lying in her bed. The story ends with a gothic depiction of sadness and loneliness, leaving the reader to guess what had happened to Homer Barron and the crime Miss Emily had most likely committed. But Faulkner made it quite apparent that the tragic events transpiring in her young life had transformed Miss Emily into the gothic figure that she later became.
While it is an extremely captivating short story to read, three features of the story make it quite daunting to digest and comprehend. First, the story is told in two time perspectives, i.e., the present and the past. Using mostly a plural first-person point of view, Faulkner narrates the tragic life of Miss Emily Grierson in a non-linear fashion using flashbacks and foreshadowing. The storyline jumps back and forth from the present time to the distant past and then back again to the present. In one scene in the past, Miss Emily is depicted as a beautiful young aristocrat greatly admired by the townspeople, whose life they are intrigued about. In another scene showing her present state, she is described as cold, eerie, and detached from reality. Many times her character comes across as dark and gothic. Faulkner rearranges the sequence of events and uses flashbacks into the immediate and distant past to show the reader how the events in Miss Emily's life influenced one another and how they collectively impacted her psychological state of mind and emotional health.
To the novice reader of English literature, Faulkner's flashbacks pose an immense problem to a clear understanding of the story's sequence of events. The reader literally has to draw out a timeline to keep track of the things happening in the story as it is often difficult to distinguish whether the event is happening in the present time frame or whether it took place in the immediate or distant past. Consider the story's opening: "When Miss Emily Grierson died, our whole town went to her funeral: the men through a sort of respectful affection for a fallen monument, the women mostly out of curiosity to see the inside of her house, which no one save an old man-servant--a combined gardener and cook--had seen in at least ten years.
It was a big, squarish frame house that had once been white, decorated with cupolas and spires and scrolled balconies in the heavily lightsome style of the seventies, set on what had once been our most select street. But garages and cotton gins had encroached and obliterated even the august names of that neighborhood; only Miss Emily's house was left, lifting its stubborn and coquettish decay above the cotton wagons and the gasoline pumps--an eyesore among eyesores. And now Miss Emily had gone to join the representatives of those august names where they lay in the cedarbemused cemetery among the ranked and anonymous graves of Union and Confederate soldiers who fell at the battle of Jefferson." In the introduction, Faulkner skillfully weaves together past and present events to establish some facts about the life and character of Emily Grierson. When the story first plays out, Miss Emily has just died, so her death occurs in the present time frame. Her house is described as a big and squarish frame house which was once grand and elegant, suggesting that Miss Emily no longer has her riches when she dies. In other words, she was once wealthy but dies a poor woman. The area in the town where she lives used to be an affluent neighborhood, but the place is now reeking with decay and degeneration. Faulkner's style of storytelling avoids the chronological order of events and invites the reader to put together the mystery and puzzle surrounding Miss Emily, piece by piece, through his temporal manipulation. The same technique of storytelling is repeated throughout the entire fiction. Hence, in order to fully understand how the story unfolds, the reader must make a mental travel back and forth through time. This creates a substantial amount of comprehension problem to novice readers of English Literature, especially those who are learning English as a second language.
The second challenging feature of the story is its use of complex language and verbose vocabulary. Although Faulkner's choice of words does wonder in enhancing the story's authenticity and emotional impact on the reader, many (especially non-native speakers of English) will likely struggle with the meaning of the expressions used, such as "an involved tale", "pallid hue", "noblesse oblige", and "hereditary obligation." These expressions require the reader to have some contextual knowledge of the American south after the Civil War to decipher their intended meanings. Furthermore, there are parts of the story the meaning of which completely eludes the reader, for instance, the following expression: "This behind their hands; rustling of craned silk and satin behind jalousies closed upon the sun of Sunday afternoon as the thin, swift clop-clop-clop of the matched team passed." This expression is so steeped in imagery that the reader must break the sentence down into small parts and evoke a mental image of what each part might mean so as to understand Faulkner's description of the scene. Readers who are unfamiliar with "craned silk and satin," "jalousies," "matched team," and "thin, swift clop-clop-clop" will need to do a considerable amount of research before they can capture the imagery in their minds and register the meaning of the entire sentence.
The third source of difficulty in understanding "A Rose for Emily" is its historical and socio-political context. The story was set in the deep cultural roots of the American south after the Civil War (1861-1865). Miss Emily was a Southern aristocrat whose family would likely have supported slavery and used black slaves on their massive cotton plantations, their primary source of wealth. Homer Barron, on the other hand, was a Northerner, a Yankee from New York, whose value system and opinions on slavery might have clashed with Miss Emily's own values and beliefs. In the American Civil War, the North (fought by Union soldiers) went against the South (fought by Confederate soldiers) to outlaw and abolish slavery, which the South permitted and perpetuated. Differences in their cultural and political roots and social standing were the reasons why Homer Barron and Miss Emily were perceived as an unlikely and incompatible couple. While Miss Emily was a respectable lady of high social ranking (i.e., an aristocrat), Homer Barron was a loud "day laborer" who often used bad language to "cuss the niggers" working under his supervision. Readers also need to know this historical background to understand why Colonel Sartoris had "fathered the edict that no Negro woman should appear on the streets without an apron," indicating that the mayor was a true Southerner and a Confederate who approved and perpetuated the racial segregation in the American south before and after the Civil War.
Hence, given these three challenging features present in the story, reading "A Rose for Emily" is no easy task. For ESL learners, it is quite impossible to comprehend the plot, characters, themes and setting without the aid of a dictionary and without knowing some historical background of the American Civil War, as well as the socio-political characteristics of the American south. Furthermore, students must have a way of dealing with its figurative language successfully as the story is replete with similes, metaphors, and imagery, not to mention the use of symbolism and personification that can easily throw the reader into a state of frustration if they fail to capture the intended meanings of these literary devices. Therefore, teachers of English Literature who must teach "A Rose for Emily" or any other similarly difficult short story, should employ engaging and appropriate instructional methods to reduce the cognitive load present in the story and render it more digestible to students.

Using Reciprocal Teaching to Teach Difficult Literature Pieces
Reciprocal teaching, coined by Palincsar and Brown (1984), is a collaborative learning activity where students read and understand a text passage or a story together in a small group of four members. Each group member plays a specific role in actively dissecting and comprehending the text. The roles include summarizing, clarifying, questioning, and predicting. The learning activity starts with the teacher (or possibly another student) reading the text passage out loud to the class, followed by the first member of the group summarizing the key ideas in the passage. In summarizing, the student (i.e., the summarizer) needs to pick out the main points or events and retells them in his/her own words to shed light on the meaning and content of the passage. The second member plays the role of a questioner and begins by asking pertinent questions on the text or story. The point of asking the questions is to increase the group's engagement with the text/story, thereby enhancing their understanding of it. Next is to have the third member act as a clarifier to provide possible answers to the questions asked and clarify the meaning of difficult words, phrases, or points. He/she can use outside examples, analogies and illustrations or consult a dictionary to clarify ideas. The fourth member is given the task of predicting what may come next in the text or story. He/she must use clues or evidence given in the text to make the predictions. In predicting, the predictor may use expressions like, "I think (this) will happen next," or "Maybe, (this character) will do or say (this) in the next scene." After completing the first cycle of summarizing, clarifying, questioning, and predicting, the learners switch roles so that every member gets the same opportunity to develop each of the four learning skills optimally. The process involved in reciprocal teaching (RT) is visually described in Figure 1.

Figure 1 Learner Roles and Cognitive Text Processing Strategies in Reciprocal Teaching
The four reading comprehension processes involved in RT are highly cognitive in nature. They require the reader to actively interact with the text, decoding and encoding the meanings of its many parts, engage in dialogue and convey and share their understanding with other group members (Oczkus, 2018). The reader's active engagement with the text helps him/her to "construct meaning from it using various kinds of background knowledge, such as linguistic knowledge of words, sentences and paragraphs, and cognitive abilities" (Carrell, 1989, as cited in Ahmadi & Gilakjani, 2012, p. 2055. Research shows that rereading parts of a passage, asking questions about it (Paris, Cross & Lipson, 1984), taking notes, consulting a dictionary to clarify difficult words, and creating summaries (Bean, 1996) are cognitive strategies that help readers deal with difficult texts. All these strategies are embedded into RT, making it a good instructional method for teaching difficult literary pieces like Faulkner's "A Rose for Emily." However, the effects of the method on English language learners' ability to comprehend highly challenging literature (i.e., short stories, poems, novella, etc.) have not been extensively explored, hence the conduct of the present study. According to Jones (2021), "reading instruction and intervention is [often] overlooked because content area teachers have many content standards they need to meet in a year… [and]…are overwhelmed with the variance in reading levels amongst students. [At the same time, teachers] do not have adequate literacy training and materials to meet students' reading needs, [but] expect students to understand texts that become increasingly challenging and specialized (p. 8).
Reading aloud by the teacher or a student Predicting Summarizing Clarifying Questioning

LITERATURE REVIEW Theory Supporting Reciprocal Teaching
Reciprocal teaching is premised on Vygotsky's social constructivist theory that heavily emphasizes the fundamental role of social interaction (dialogue) in the development of students' cognition and language use (Foster & Rotoloni, 2005). He described the art of teaching and learning as "much more than face-to-face interaction or the simple transmission of prescribed knowledge and skills" (Daniels, 2001, p. 2). Vygotsky's idea of teaching places much emphasis on dialogue (between teacher and students and between students and students) and collective construction of knowledge that is done in a social group (Wells, 1999).
Consistent with Vygotsky's beliefs, RT as an instructional method actively employs cognitive text processing strategies and has them played out in a social interaction process. It uses thinking aloud, questioning, discussion of ideas and shared, negotiated meaning through four reading roles to systematically develop students' text comprehension. At the same time, the four roles-summarizing, questioning, clarifying, and predicting-collectively constitute a guided reading strategy that parallels Vygotsky's idea of scaffolding, defined as a variety of instructional techniques or learning activities used to move students progressively toward a firmer content understanding (Erbil, 2020).
In reading a short story, asking questions helps students to think about the literature and invites them to discuss it at a deeper level via summarizing the plot, theme or setting, clarifying difficult words or ideas, and predicting what might happen next in the storyline. As such, the strategy utilizes question-answer exchanges which require a lot of language use that actively works to develop students' cognition and metacognition, an interactive social process that is reflective of what Vygotsky said learning and meaning making should be. In a typical ESL classroom, much of this process is controlled by the teacher, but in the collaborative reciprocal teaching set-up, the control of this process is left in the hands of the students, rather than the teacher. Over time, the method will not only develop students' thinking, metacognition, and language use, but will also empower them with the lifelong ability for independent learning, which is one of the many ultimate goals of literacy education.

Effects of Reciprocal Teaching on Reading Comprehension
Scores of text processing studies have consistently demonstrated that explicitly teaching students how to read a text for meaning-by identifying and communicating the main ideas in it, either orally or in writing, summarizing the major points, and creating questions based on the text-significantly improves their reading comprehension (Ahmadi & Gilakjani, 2012;Alharbi, 2015;Hughes, Scales, & Scales, 2021;Joseph, Alber-Morgan, Cullen & Rouse, 2015;Maplethorpe, Kim, Hunte, Vincett, & Jang, 2022;Shokrpour, Sadeghi, & Seddigh, 2013). The positive results of these studies lend support to the strategies embedded in reciprocal teaching, particularly summary writing or summarizing, asking or creating questions and talking about the text, vis-à-vis predicting and clarifying ideas. The main purpose of reading is to construct and derive meaning from written text, and this act of constructing meaning, according to the Texas Educational Agency (2012), should be interactive, strategic, and adaptable. Interactive means the reader must actively process the text and exchange his/her thoughts about it with the peers in his/her social group. Strategic means that the reader must have a clear purpose for reading the text and use a variety of meaning making strategies to achieve the purpose. This relates directly to the four strategies used in reciprocal teaching. Lastly, "adaptable" means that the reader may change or modify their reading comprehension strategies according to the type of text being read and its degree of complexity. This aspect is reflected in the addition of the "reader" role to the existing four RT strategies.
The empirical evidence on the effectiveness of reciprocal teaching as a whole has been mixed and inconclusive. Much previous research confirms that RT is an effective reading strategy that significantly promotes reading comprehension (Ahmadi, 2016;Choo, Eng & Ahmad, 2011;Hamdani, 2020;Huang & Yang, 2015;Koch & Spörer, 2017;Kula & Budak, 2020;Pilten, 2016), particularly for struggling readers (Cockerill, O'Keeffe, Thurston & Taylor, 2022;O'Hare et al., 2019;Thurston et al., 2020) and EFL learners (Navaie, 2018;Rojabi, 2021). Rawengwan and Yawiloeng (2020) went further to demonstrate that RT even improved EFL learners' metacognitive ability, in addition to increasing their comprehension of the reading text. RT is effective because it "encourages a slower pace of reading to nurture thinking about the text…[which] facilitate[s] a more in-depth understanding of the text" (Cockerill et al., 2022, p. 6). In Dew, Swanto and Pang's (2021) metaanalysis of 18 quasi-experiments, RT was found to exert a statistically significant impact on learners' reading comprehension scores in a majority of the studies (n = 14).
However, a substantial body of evidence stemming from other studies is suggesting quite the opposite. For instance, in Dew et al.'s (2021) meta-analysis, four studies reported mediocre results with roughly a 50% success rate. Likewise, Kula (2021) discovered that RT did not have any significant effect on 2nd graders' reading comprehension efficacy, attributing the lack of positive results to the subjects' being very young learners and unable to benefit from the highly cognitive activities of summarizing and questioning, especially. Although his qualitative data showed that both students and teachers felt that RT had significantly and positively improved the former's reading comprehension, Kula's (2021) quantitative results did not support this assertion. According to Rosenshine and Meister (1994), RT is more effective for students in the older age groups and learners with poor reading comprehension skills. They concluded this based on the results of a meta-analysis of 16 quantitative studies. Muijselaar et al. (2018) claimed that RT only increased learners' awareness and knowledge of reading strategies, but failed to improve reading comprehension directly, linking the negative results partly to the subjects' young age and inappropriate use of the strategies. In Van Keer and Verhaeghe (2005), no effect was found on the perceived reading comprehension self-efficacy of 2nd year university students.
A central principle of RT and other forms of collaborative learning is that the text assigned to students must be at a level that can be effectively understood and shared by them, meaning that it should not be too easy or too difficult for students to master on their own. But when the text is a difficult one, appropriate support and feedback must be given to facilitate learning during the reciprocal teaching activities (Oczkus, 2010). This notion has not been tested with ESL learners in their dealing with a challenging and daunting literary piece such as the short story "A Rose for Emily." Furthermore, to date, there has been very little research on RT that involves Nigerian ESL learners. This had warranted the conduct of the present study.

Research Objective and Question
The aim of the experiment was to demonstrate the effects of a strategic RT intervention on Nigerian ESL learners' mastery of "A Rose for Emily," a very difficult early 20th century American short story written by William Faulkner. Mastery was measured via students' test scores in recall, comprehension, and vocabulary. The research question asked was, "What are the effects of RT on Nigerian ESL learners' mastery of A Rose for Emily in terms of their recall and comprehension of the text and understanding of its vocabulary?"

Research Design
This research was a true experimental study following the pretest-posttest control group design that can be schematically described as follows (Figure 2):

Figure 2 Pretest-Posttest Control Group Design with Randomization
The experiment started with the administration of a pretest (O1 and O7) containing 135 questions that established the subjects' prior knowledge of the short story. The subjects were then randomly assigned (R) to either the treatment or control group based on their matched pretest scores (i.e., matching). Five RT lessons (X1 to X5) and five direct teaching sessions were conducted for the treatment and control groups, respectively. At the conclusion of each lesson, the subjects took a posttest, resulting in five small posttests altogether, i.e., O2 to O6 for the treatment group and O8 to O12 for the control group. Randomization (R) was exercised when the subjects were randomly selected into the treatment or control group based on their performance in the pretest.

Population and Subjects
The ideal population was all 3rd year university students specializing in English at all colleges of education in Nigeria, while the target population comprised 300 3rd year ESL students doing their English major at a selected state-owned college of education in the Lagos metropolis. There are about 85 colleges of education in Nigeria, which consist of one military, 21 federal, 42 state-owned, and 21 private colleges, all of which have been accredited by Nigeria's National Commission for Colleges of Education (NCCE). Colleges without the NCCE accreditation were excluded from the population.
The subjects were 60 3rd year (300 level) college of education students who, at the material time, were attending a randomly selected college of education in Lagos. They were 39 males and 21 females aged between 18 and 20. All subjects spoke at least one of Nigeria's three main languages, i.e., Yoruba, Hausa, and Igbo. At the material time, they were learning English as a second language (ESL) at the selected college of education. In Nigeria, English is not just a subject to be learned in schools; it is also the medium of instruction in Nigerian education from preschool to the tertiary level. Hence, the subjects of this study had been receiving instruction in English for at least nine years of their schooling.

Subject Selection
The pretest was administered to all 3rd year ESL students in the sampling frame, i.e., the target population (N = 300). Based on their performance, 60 students with the least prior knowledge on the short story (i.e., those with the lowest pretest scores) were identified and shortlisted. Students with very high scores, i.e., higher than 30 points (or 70% of the total pretest score), were removed from the list, resulting in a substantial number of students being excluded from the study. From the finalized list of 60, the students' scores were paired in a matching process. For example, Student A with 5 marks was paired with Student B who also obtained 5 marks on the pretest. One of them was then randomly placed in the treatment group, while the other, in the control group. This process was called randomization and it continued until both groups had 30 subjects in each. At the end of the process, two equivalent groups were obtained. The rationale for this matching process was to ensure that the treatment and control groups were as identical as possible in terms of their prior knowledge before the intervention, thereby reducing the likelihood of the variable contaminating the results.

Reading Material for the Subjects
The reading material assigned to both groups was the short story "A Rose for Emily," which had five parts to it. The parts were turned into five individual lessons delivered via reciprocal teaching to the treatment group and direct instruction to the control group. The five lessons are outlined in Table 1. An excerpt from Part One of the story is shown in Figure 3 below.

Lesson Plans for the Teachers
Two sets of lesson plans were prepared for the teachers to guide their implementation of RT and direct instruction in the experiment. The first set contained five RT lesson plans prepared for the teacher teaching the treatment subjects. The second set of plans comprised the lesson blueprints for the teacher teaching the control group via direct instruction. Both sets of plans (10 of them altogether) were equal and the same in all aspects-i.e., the content, learning outcomes, materials, and time-except for the manner in which the content would be taught. The teachers had the opportunity to test and practice their teaching plans before actually using them in the experiment.

Intervention: Reciprocal Teaching
The intervention came in the form of five sets of reciprocal teaching activities conducted by EFL learners in the treatment group. The class was broken into roughly six groups with five members in each group and was asked to study "A Rose for Emily" by first reading it aloud, and then summarizing the key points, followed by asking questions, clarifying muddy points and difficult vocabulary, and finally by predicting future events in the story. To facilitate these cognitive tasks, the groups were each given a strategy packet containing index cards with four student roles: predictor, summarizer, questioner, and clarifier. Each card showed the subjects how to play their roles successfully, that is, how to summarize, predict, question, and clarify the events in the short story in the expected manner. The cards were adapted from the strategy packet developed by an Australian teacher (Reading Rockets, 2019), with the fifth role of reader being added by the researchers to the original four roles.
Since the story was a challenging one, the subjects were instructed to read it in small chunks and dissect it paragraph by paragraph via RT. For instance, Part One was broken into eight small sections, which means that eight rounds of RT were performed by the subjects to read and comprehend just the first part of the short story. For every round, the subjects were asked to switch roles so that everyone got to be in all four roles. The same standard pattern of learning the short story, with the subjects' switching roles with every round, was implemented across the five lessons.

Instruments
Two main instruments were used to assess the effects of RT on the subjects' mastery of "A Rose for Emily" in terms of recall, comprehension, and vocabulary. They were a set of five mastery tests on the short story, and a set of scoring rubrics.

Mastery Tests
The researchers developed five tests worth 135 points to measure the subjects' mastery of the story using open-response questions. Each test, worth 45 points, had three parts. The first part measured the subjects' recall of factual information presented in the story, such as "Who was Miss Emily?", "Who was Tobe?" and "What type of house was Miss Emily's house?" The second part measured their mastery of the difficult vocabulary used in the story, while the third part tested the subjects' understanding of the story's plot and meaning. All three constructs, i.e., recall, comprehension, and mastery of vocabulary, totalled 45 marks each. Accordingly, the test came up to a total of 135 marks. Higher scores on the tests meant greater mastery of the short story. The breakdown of test items by type and total score is shown in Table 2.

Scoring Rubrics
Five scoring rubrics-one for each test-were used to grade the students' answers to the openresponse questions measuring recall, comprehension, and vocabulary. The rubrics were selfdeveloped, detailed blueprints that provided correct and acceptable answers to each question in the tests. They were validated by two subject matter experts. The reason for creating the rubrics to be as comprehensive as possible was to increase the accuracy and consistency of the grading process by two independent test scorers. This procedure helped the study to obtain standardized test scores on recall, comprehension, and vocabulary from the two scorers, hence establishing the study's test reliability. The scoring rubrics explicitly stated the number of points that should be awarded for each answer, hence reducing errors in the scoring, which consequently increased the reliability of the data indicating the subjects' mastery of the short story. A sample scoring rubric for one part of the tests measuring recall is given in Table 3.

Validity of the Tests and Scoring Rubrics
The self-developed tests and scoring rubrics were submitted to two public university lecturers for content validation. Both experts were knowledgeable about the short story and the constructs being examined (i.e., recall, comprehension, and vocabulary mastery). They were requested to review the contents and verify four things, namely (1) the alignment between the test items and the contents of the short story; (2) the accuracy of the test items in measuring either recall or understanding according to how the constructs were defined in Bloom's revised taxonomy (Krathwohl, 2002); (3) the accuracy of the answers provided in the scoring rubrics; and (4) the appropriateness of the marks distribution for correct answers. Both experts confirmed the content validity of the test items and scoring rubrics with minor adjustments. They also verified the question formats as appropriate for measuring the three constructs.

Reliability of the Tests and Scoring Rubrics
The tests and rubrics were pilot tested to obtain information on their practical aspects, such as item clarity and amount of time required to complete the tests. Thirty (30) ESL students from the same college's English department who were already familiar with the story were asked to take all five tests. Their responses were graded by one of the two teachers who would later be involved in the study. The teacher was instructed to follow the scoring rubrics closely in grading the answers. The scores were then subjected to a reliability analysis using the Cronbach's alpha. Table 4 shows the results of the analysis, i.e., the reliability estimates of the data derived from the tests used. Vocabulary .825 Overall Mastery .804 The results indicated good reliability for vocabulary and overall mastery, but rather low estimates for recall and comprehension, although any value of 0.60 and above is considered acceptable for social science research.

Experimental Procedures
Two female ESL instructors from the college of education were employed to teach the short story to the treatment and control groups. The lessons were taught concurrently by them over five weeks. The first teacher was assigned to the treatment group and taught the five lessons using reciprocal teaching, while the second teacher taught the same five lessons to the control group using direct instruction. Each teacher's implementation of the lessons was guided by the lesson plans created by the researchers. The teachers spent the first week of the experiment to practice their respective instructional strategies, i.e., RT for the treatment group and direct instruction for the control group, using a different short story. The actual experiment began in the second week. In the first meeting, the teacher first modelled the RT strategy to the treatment subjects. After obtaining some understanding of the strategy, the subjects tried out the four roles in their groups.
Each lesson took two hours to conduct for both groups. The first 80 minutes were spent on teaching and learning the short story, while the remaining 40 minutes were spent on the posttest. In every session, the teachers explicitly stated the objectives of the lesson and distributed the necessary reading materials. To ensure no experimental bias and reduce teacher effects on the outcomes, the teachers were given a daily script that provided detailed instruction on what to say and how each lesson should be carried out according to their respective strategies.
Students in the control group learned the short story via direct instruction. They were first asked to read the short story aloud and then engaged in a silent reading of it. The teacher explained the plot and characters and engaged in the regular question-and-answer session with the class, pausing at critical key points in the story to check on students' comprehension of the plot. She also asked questions regarding their understanding of the vocabulary and illustrated their meaning using the whiteboard. Each lesson was closed with a summary of the content by the teacher.
Students in the treatment group received the reciprocal teaching intervention and studied the short story in groups of five, with the added role of a reader. The RT activity began with the first student reading the first paragraph of the story aloud to the group members, followed by the second student (i.e., summarizer) giving a summary of the key points and the third student (i.e., questioner) asking a set of questions. The fourth student (i.e., clarifier) then provided answers to the questions posed and clarified ideas to the group. The fifth student ended the first cycle of learning by making a prediction of what might happen next. The students used the strategy packet containing index cards that showed how to summarize, ask questions, clarify, and predict to help them execute their roles successfully. The roles were switched and rotated with every learning cycle. During this process, the teacher acted as facilitator, going round the class to see if the roles were done correctly.

Data Collection and Research Ethics
Approvals for data collection were formally obtained from the relevant offices in the selected college of education in Lagos, as well as from the Nigerian College of Education Board. Students were then duly informed about the research and were asked to sign a consent form upon their willingness to participate the study. Parents of the willing students were also given a consent form to sign. All parties were made to understand that they were in no way forced or obligated to participate in the study. Data collection began after every necessary documentation, like official letters and consent forms from the parents and students, was received in full.
The two female teachers assigned to teach the groups were given a consent form to sign, saying that they agreed with the research terms and conditions. The consent form included a confidentiality agreement which contained the prohibition of data disclosure. The teachers were also trained to use their respective strategies. Both watched a video on reciprocal teaching and its classroom implementation and another video on direct instruction before meeting on a Skype call with the researchers to discuss the experimental procedures. In the online discussion, the researchers further explained the concept of reciprocal teaching to the teachers and later emailed them the materials they needed to study for the short story, explaining as well how the materials should be effectively used. Most importantly, the story's plot, symbolism and meaning were adequately communicated to the teachers. These contents were shown on a Power Point presentation during the Skype call. Finally, the teachers were instructed to review their respective lesson plans and teaching materials prepared for the experiment prior to teaching the short story in class.

Data Analysis
The subjects were awarded 1 point for each correct answer, and zero for every incorrect answer. The total score for each construct covered in each test was 27. Their total posttest scores from the five sessions were recorded for recall, comprehension, and vocabulary mastery, from which the subjects' learning gains were computed (i.e., posttest minus pretest). Descriptive statistics and independent samples t-tests were run on the data to determine the effects of RT on recall, comprehension, and vocabulary mastery. Cohen's d effect sizes were computed on the significant differences found between the treatment and control group's mean scores in terms of recall, comprehension, vocabulary, and overall mastery of the short story. Table 5 shows the group statistics of the subjects' performances in the pretest and posttest and their gain scores. Based on their mean scores in the pretest, it can be seen that both groups were on an equal footing before the intervention, with only one-point difference between them in overall mastery, in favour of the control group (M = 50; SD = 14.71) over the treatment group (M = 49; SD = 14.44).

Descriptive Analysis of Gain Scores
For recall, after the intervention, the treatment group increased by 24.1 points from an M = 17.0 (SD = 5.61) to M = 41.1 (SD = 2.72), while the control group increased by 15.3 points, moving from M = 17.1 (SD = 5.21) to M = 32.4. The treatment group's gain score (24.1 points) exceeded that of the control group (15.3 points) by 8.8 points. For comprehension, the treatment group increased by 19.0 points compared to 11.8 points by the control group. For vocabulary, the treatment group increased by 24.5 points, while the control group increased by 17.8 points.
In terms of overall mastery of the short story, the treatment group had the upper hand with a 67.6-point gain score against a 44.9-point gain score by the control group.

Effects of RT on ESL Students' Mastery of the Short Story: Independent Samples t-Test Results
The groups' gain scores were subjected to four independent samples t-tests to determine if differences in mastery between the treatment and control groups were statistically significant. For recall, the difference in gain scores was significant and in favour of the treatment group (M = 24.10, SD = 6.35) over the control group (M = 15.30, SD = 5.55), t(58) = -5.712, p = .001. The effect size of the difference was large at Cohen's d = 1.47, suggesting a very large magnitude and practical importance. The results were similar for comprehension. A statistically significant difference was found between the gain score of the control group (M = 11.77, SD = 5.43) and that of the treatment group (M = 19.03, SD = 4.92), t(58) = -5.430, p = .001). The effect size of the difference was Cohen's d = 1.40, also suggesting a difference of a large practical importance.
For the subjects' mastery of vocabulary, the t-test results pointed to a statistically significant difference in the gain scores of the control group (M= 17.87, SD = 6.47) and the treatment group (M = 24.47, SD = 6.81), t(58) = -3.848, p = .001, also in favour of the treatment group. The effect size of the group difference was recorded at Cohen's d = 0.99, similarly suggesting a large impact of the RT intervention, albeit lower than that for recall and comprehension. In terms of overall mastery, the t-test results showed the treatment group (M = 67.60, SD = 11.91) significantly outperforming the control group (M= 44.90, SD = 8.55), t (58) = -18.45, p = .001) at Cohen's d = 4.76. These results suggested that RT produced a large practical impact on Nigerian ESL students' ability to master a difficult literature piece through their own independent collaborative study of it.

DISCUSSION
As indicated in the review section, the existing empirical evidence in the large body of research literature on reciprocal teaching appears to be inconclusive, with some results pointing in the positive direction, while some others, in the opposite way. Theoretically, RT is expected to improve learners' cognition, independent learning skills and overall learning outcomes, due to the fact that its activities of summarizing, questioning, clarifying and predicting galvanize the minds of students and encourage them to be actively engaged in dissecting the text. The four reading activities provide learners with what Vygostky termed as scaffolding, a sort of framework that guides learners toward a better understanding of the text being read. In addition, having to do the reading and dissecting of the text in a social group means that learners have to talk about the text and its content. The exchange of ideas, thoughts and reflections in the social group further deepens learners' comprehension and interpretation of the text. The positive results of the study, with their large effect sizes indicating huge practical impacts of RT on student learning and content mastery, appear to support the theoretical notions underpinning the strategy.
The subjects' large gain scores on all four measures of the dependent variable (i.e., recall, comprehension, vocabulary, and overall mastery of the difficult short story) and statistically significant t-test results in favour of the treatment group corroborate the works of Choo et al.  Thurston et al. (2020). In particular, the results support the contention that using RT for literature study is beneficial for EFL learners, as demonstrated in the works of Navaie (2018); Rojabi (2021); Rawengwan and Yawiloeng (2020); Cockerill et al. (2022) and Dew et al. (2021). In fact, Cockerill et al. (2022) asserted that RT works effectively because learners can do the processing and dissecting of the text in small chunks, at their own pace, within a peer group that they feel comfortable with. Psychologically, students learn better when they feel safe in the social group they are assigned to.
The large impacts of RT, as indicated by the effect sizes, suggest that ESL teachers and instructors should start taking this instructional strategy seriously. It has proven to be a viable solution to helping ESL learners read difficult text in the English language. They should begin thinking of ways to incorporate the strategy into the English language curriculum to improve learners' thinking and reasoning, as well as their ability to share ideas and communicate their thoughts convincingly. The strategy also helps learners to approach the study of a difficult text in small bites, question ideas in the text and put forward an argument for text refutation. Thus, ESL teachers and instructors should use RT side by side with direct instruction to complement traditional teaching and increase learners' ability to comprehend difficult literary texts.
The results negated Muijselaar et al.'s (2018) claim that RT could not improve reading comprehension directly, as the reading comprehension of the ESL learners in this study did indeed improve markedly. Since the subjects were tertiary-level students, the study could not fully corroborate Rosenshine and Meister's (1994) contention that RT is more effective for students in the older age groups than it is for young learners as it has no basis for verifying the claim. This would require another research undertaking.

LIMITATIONS AND FURTHER RESEARCH
The results are constrained by three limitations. First, despite the rigorous measures taken to establish the reliability of the data and mastery tests (i.e., by content validating the questions and preparing explicit scoring rubrics), the inter-rater correlation coefficients for recall and comprehension were rather low, falling below the threshold of .70 for social science research. This necessitates that the results suggesting the impacts of RT on recall and comprehension be read and interpreted with some caution. Future studies interested in using the mastery tests on "A Rose for Emily" developed in this research must re-examine the items and rubrics and subject them to another round of validation and pilot testing before they can be used to measure the two constructs reliably.
Second, two teachers-instead of just one-were employed to teach the treatment and control groups. This might have created variations in instructional styles and quality and introduced systematic bias that could have confounded the internal validity of the results. In future experimental studies, researchers should be very careful in addressing threats to the internal and external validity of their experiments. Third, the study had neglected to investigate the subjects' views of RT as a strategy for learning literature. Hence, the researchers have no way of knowing which aspect(s) or RT or which student roles were most effective in helping them comprehend the short story. Interviewing the subjects, either individually or in small groups, would have afforded the study with great insights into which specific features of RT had been instrumental in reducing the complexity of a difficult short story. In conclusion, future experimental studies should not neglect to incorporate a qualitative element into their research design to acquire more insightful data on what helps ESL learners use RT to learn literature better.