Effect of Differentiated Instruction on Linguistic and Rhetorical Features of Narrative Writing Performance across Different Levels of Proficiency

Research in English Language Pedagogy

مقاله 10، دوره 7، شماره 2 - شماره پیاپی 13، مهر 2019، صفحه 405-426   

نوع مقاله: Original Article

شناسه دیجیتال (DOI): 10.30486/relp.2019.665954

نویسنده

Hadis Toofani Asl^*

Department of Foreign Languages, Shanghai University. Shanghai, China.

چکیده

This study intended to explore the linguistic and rhetorical features of written narratives produced by language learners of different levels of proficiency. A large and comprehensive number of linguistic and rhetorical features were analyzed adopted from Hinkel (2002). Learners from a mixed-age group in a private language institute in Iran took part in this study. Three levels of B1, B2, and C1 learners were assigned into two conditions: Differentiated Instruction (DI, N = 84) that received individualized writing instruction according to each writer’s needs and skills and Traditional Instruction (TI, N = 76) that presented the features followed by some exercises. Results revealed that both C1 and B2 learners used more accurate linguistic features, while only C1 learners were successful in employing rhetorical features. In TI, however, C2 learners implemented a restricted number of linguistic and rhetorical features. Significant differences were found between DI and TI groups in each proficiency level. A comparison of the effect sizes, however, indicated that DI was less effective in the case of rhetorical features.

کلیدواژه‌ها

Differentiated instruction؛ Level of proficiency؛ Linguistic features؛ Narrative؛ Rhetorical features

اصل مقاله

1. Introduction

In spite of the plethora of studies on oral narratives in the field of second language acquisition, comparatively lesser attention has been devoted to written narratives produced by foreign language learners although this genre is a significant area of query both in language teaching pedagogy and in the assessment of Foreign Language (FL) competence (Kormos, 2011). Narratives belong to one of the most frequently instructed written text types in general language teaching courses commencing from the beginning level, and they are also entailed in the curriculum of writing classes for higher level college learners. Furthermore, various high-stakes language proficiency tests also implement narrative tasks to measure FL learners’ writing competence (such as all levels of Cambridge ESOL proficiency tests). Besides, narrative accounts of events are crucial in authentic written interactions, and narratives can be included in different text types, such as providing evidence in argumentative writing.

At all levels of education, difficulties with writing have an adverse effect on academic achievement as writing is a requirement of most curriculum areas (Parr & Jesson, 2012). Writing is a significant means to advocate learning and cultivate knowledge, as well as to evaluate performance; thus, the reason for the importance of learners being able to express themselves clearly and coherently in writing (Kiuhara, Graham, & Hawken, 2009). Improving the writing quality of FL writers has, therefore, potentially far reaching effects not only for their academic achievement at school, but also for their occupational prospects.

Below, the importance of narration for FL learners in terms of expressing meaning better and the distinct instructional practices to foster this capability are presented.

 

2. Literature Review

2.1. Meaning Making Through Narratives

The tradition of telling and re-telling stories, both to oneself and socially, has been a commonplace practice throughout history. Within the positivist tradition, this practice has largely become side-lined as folk-knowledge that has little or few implications for ‘scientific’ understanding or clinical practice (Clandinin, 2006). By their very nature, people lead storied lives and tell stories about their lives as a way to engage with and interpret the world, in order to make it personally meaningful and relevant (Riessman, 2008). On closer inspection, it becomes apparent that human experience is typically narrated as ideas; therefore, experience cannot be reduced to statements or definitions without losing its richness and complexity. The stories or narratives that are created provide a way to organize and understand experience and to render it personally significant (Polkinghorne, 2007), without narratives life would essentially be a set of disconnected events with little connection to the past or present. The meanings and understandings that narratives give have important implications for the way people live their lives, and as such they are an important source of information in terms of getting a full, rather than two dimensional, and picture of the people or groups under study.

Narrative research is an attempt to value and understand the stories that people tell (Lieblich, Tuval-Mashiach, & Zilber, 1998). It assumes that people organize their life experiences into stories that the stories told depend on the individual’s past and present experiences, and that stories are overlaid by multiple voices (Riessman, 2008). In doing so, it strives to preserve the richness and complexity of human experience and to locate this experience firmly within context. Whilst each individual makes meaning in their own way, the process of meaning making also occurs within a social, cultural, and historical context (Riessman, 2008). The stories people tell are seen as being dependent on the range of stories available within their social and cultural context. Indeed, what constitutes a story in one society or culture could constitute something quite different in another (Mendieta, 2013).

Therefore, narratives as a genre to elicit individual inclination towards producing second language (L2), provides researchers an opportunity to get insights about FL/L2 learners’ capacities in expressing the meaning using their own individual views (Khezrlou, 2019). However, as individuality comes into play, narratives are not considered to be an easy way to articulating language particularly for limited level FL learners. There are different instructional practices that can support the development of narratives in different ways.

 

2.2. Teaching as Inquiry Approach

In the Teaching as Inquiry Cycle (TIC), learner accomplishment data “drive decisions about instruction, scheduling, and interventions [as teachers] gather and analyze real-time data from formative assessments to inform instruction and to target remediation targeted interventions” (Calderon et al., 2011, p. 117). Figures 1 illustrates the cyclic nature of TIC where assessment and feedback is equivalent to informing teachers on proper teaching content and differentiated teaching practices to ascertain consistent progress, interest, motivation and self-regulation (Harks, Rakoczy, Hattie, Besser, & Klieme, 2013).

Research corroborates that differentiation provides amendment to the learners’ writing quality and that a one-size-fits-all type of teaching does not lead to the availability of learning content, nor does it lead to advanced learning for all learners (Watts-Taffe, Laster, Broach, Marinak, McDonald-Connor, & Walker-Dalhouse, 2012). Differentiated instruction (DI) “responds to human diversity and aims for cognitive flexibility... and make use of variability, rather than implement uniform techniques or routines” (Darling-Hammond & Snyder, 2000, p. 524). It has, therefore, the potential to increase learning for all learners as curricula, teaching practices, resources, learning activities and student products are changed in order to meet individual needs (Santengelo & Tomlinson, 2012).

 

 

Figure 1. Differentiated instruction.

 

Diagnostic teaching constitutes the core of adaptive teaching (Houtveen, Booij, de Jong, & Van de Grift, 1999). Influential differentiated teaching necessitates teachers to know their students as the learners because it is “virtually impossible to make content relevant for learners whom you don’t know” (Littky, 2004, as cited in Tomlinson, 2008, p. 27). Teachers need to therefore specify and respond to individual learners’ writing needs, interest, and learning preferences, which would result in the needs of all learners in diverse classes. Learning takes place when learners are pushed into their zone of proximal development (ZPD) by material being practiced and tasks being set at a slightly more difficult level than they can deal with on their own. When material and activities are below their achievement level, “no growth will occur. In like manner, [when they are] well above their zone, children will be confused and frustrated” (Joseph, Thomas, Simonette, & Ramsook, 2013, p. 29).

Differentiation can target content (i.e., the subject of learning), process (i.e., the way of learning), or product (i.e., the output of learners’ mastery of the knowledge or skills; Huebner, 2010; see Figure 2). Content differentiation refers to each learner being instructed the same curriculum, yet that the content is distinct quantitatively or qualitatively. It brings about modification in content without ignoring the curriculum to which all children are entitled and demand teachers to either change or adjust content (Joseph et al., 2013). These comprise the changing of scaffolds, and the degree of complexity of material, as well as corresponding activities, teaching purposes, and objectives to high-stakes standardized tests (Santamaria, 2009). Learners with content knowledge above their Curriculum Level need to be extended, whereas those below need to be more prosperous with a smaller amount of content.

 

 

Figure 2.Learning cycle and decision factors in differentiated instruction.

 

In order to optimize the learner’s progress, teachers are suggested to integrate different ways to teach writing using feedback to clarify the tasks that need to be undertaken and the goals behind them and to inform the learners about their progress (Hattie & Timperley, 2007). Feedback thus makes learners aware of what good performance is, how their performance relates to good performance and what they should do to bridge the gap between the two (Sadler, 1989). This form of feedback is referred to as Process Oriented Feedback (POF) or formative feedback and it is continuous and merged into the whole writing experience (McKensie & Tompins, 1984). POF is an individual reference standard with the goal of removing the gap between an individual’s real and desired level of performance (Harks et al., 2013). Feedback directed towards the outcome of learning is considered as Grade Orientated Feedback (GOF) or summative feedback (Harks et al., 2013). GOF is socially referenced with the intention of comparing individual performances with that of peers and is usually scored as a grade, for instance, the primary school grading system in Iran is like Not-Achieved, Achieved, Merit, and Excellence. Lipnevich and Smith (2009) investigated 464 college students working on an essay examination to examine the influence of types of feedback and praise on performance. They concluded that the highest-performing group in the study was the one receiving detailed feedback that was provided by the instructor with no grade and no praise.

The following research questions were in focus in this study:

1. How does the written narrative performance of different level FL writers differ in terms of linguistic characteristics after receiving differentiated writing instruction?
2. How does the written narrative performance of different level FL writers differ in terms of rhetoric characteristics after receiving differentiated writing instruction?

 

3. Methodology

3.1. Design of the Study

This study had a posttest only quasi-experimental design since there was no random sampling and participants’ narrative performance was measured following the posttest. The type of treatment, namely traditional or differentiated instruction, as well as the learners’ level of proficiency were the independent variables. Learners’ narrative performance in terms of linguistic and rhetorical features was the dependent variable.

 

3.2 Participants

The Iranian participants were learners (N=160) from six intact classes at a private language institute. Three classes of B1 (N=26), B2 (N=30), and C1 (N=28) level of proficiency learners were exposed to differentiated instruction and the other three classes of B1 (N=23), B2 (N=26), and C1 (N=27) level learners received traditional explicit instruction. These levels of proficiency were opted for in order to measure the impact of instruction on the consolidation of already noticed structures in the target language. Learners’ age was between 16 and 24 years and they were all female. Participants’ level of proficiency was measured through Preliminary English Test (PET).

 

3.3. Instruments and Materials

Picture-cued written narrative tasks adopted from Heaton (1975) were used in this study. The tasks were: A surprise, John and his boxes, Picnic, A tiger in the mountains, The chase, Football task, Wet paint, The grass, The monkey, The thief, Elephant in the circus, and The story of a dog.These cartoons have similar structures (six frames and a clear punch line) and are presented in the correct order and made a coherent story line. The input to the task was presented visually and the instructions were given in the first language of the participants, namely Persian. This type of task did not demand the conceptualization of the plot and was therefore deemed to reduce cognitive load on the participants with respect to conceptualizing their meaning. This task, however, could potentially extract the use of particular low-frequency words such as carriage. Learners were given 2 minutes to plan their narratives. It needs to be noted that all of the tasks were also rated by a second rater (a PhD candidate in Applied Linguistics with over 5 years of teaching experience). The inter-rater reliability estimates using Pearson Correlation Coefficient are as follows: A surprise (r=0.98), John and his boxes (r=0.99), Picnic (r=0.95), A tiger in the mountains (r=0.97), The chase (r=0.98), Football task (r=0.96), Wet paint (r=0.96), The grass (r=0.99), The monkey (r=0.97), The thief (r=0.99), Elephant in the circus (r=0.95), and The story of a dog (r=0.94).

 

3.4. Data Collection Procedures

A total of 12 narrative tasks were used in this study. Learners were given the choice to select one out of two narratives in each treatment session. The elements of writing approaches were used such as explicit instruction (The Product Approach), the use of models (The Product Approach and the Genre Based Approach), and writing in recursive stages (The Process Writing Approach). Writing was taught within an overarching process writing approach with recursive stages (planning, drafting, and revising). Learners were equipped through pre-writing tasks with the proper content and genre knowledge for the tasks. Collaboration between learners to improve content knowledge and narrative accuracy was encouraged by learners discussing and proofreading each other’s work. In the prewriting stage, learners usually worked in groups or pairs to brainstorm ideas prior to individually writing ideas on paper. The learners were initially encouraged to only focus on content, and putting their ideas on paper. The reasoning was that a focus on accuracy could become a blockage and inhibit learners from expressing their ideas. Once their work was drafted, learners proofread their own and each other’s work, where after they individually wrote their final copies.

Intentional moves of teaching to meet learners’ writing needs was the next stage and comprised deliberately teaching writing strategies and processes, grammar, rhetoric, syntax, and structure. Published models and exemplars of other learners’ essays were used to exemplify features of the narrative genre, and explain the quality of narratives with the level of achievement. The teacher taught her learners proofreading skills through first explaining what proofreading is, modeling it, and then letting them practice finding errors in a text. The teacher integrated intentional linguistic and rhetoric instruction in writing lessons on paragraph development. For practice, learners were asked to correct paragraphs riddled with grammatical errors, after which they discussed both the components of paragraph writing and the linguistic and rhetorical errors.

The teaching as inquiry approach was evident in the specification of background knowledge to ascertain new knowledge was constructed based on existing knowledge with differentiating teaching, and content chosen according to learners’ needs, pace, and learning profiles. This was accomplished in the following way: At the start of the year, the teacher analyzed their learners’ first piece of narrative writing to identify each learner’s unique writing strengths and weaknesses. The teacher attempted to address learners’ difficulties in small groups or individually, and not often through whole class teaching. For revision purposes, the teacher addressed difficulties that arose through mini-lessons to the whole class instead of merely those it pertained to. As well as differentiating the instruction learners were exposed to, the teacher also made allowances for learners working at different paces. With respect to working at different paces, learners had started to work on a more individualized program early in the year. The teacher started the lessons by verbally giving obvious and differentiated lesson purposes and instructions. The instructions to learners were also displayed on the screen as follows: (a) If you have completed your writing already: Proofread your work. Make sure it is as accurate as possible. Check the accuracy of the language you have used. (b) If you have started the writing but have not completed it yet: Complete the writing, use the notes that were provided to you, but keep in mind you are allowed to include any information that you think helps to describe your point. Make sure to narrate the events well and in order in your writing.

(c) If you have not yet started your writing: Use the notes and start. To be able to improve the most important thing is to start.

The instruction was also differentiated by giving learners an element of choice regarding narrative topics, and thereby allowing them to use their background knowledge, strengths and interests.

Continuous and clear process feedback was given. The teacher regularly offered verbal, written and peer feedback on both linguistic and rhetoric features to help learners improve their narrative quality. Teachers had conference with each learner to provide individual feedback, discuss progress, and establish next steps.

 

3.5. Measures of Linguistic and Rhetoric Features

A wide range of linguistic and rhetorical features was prioritized since there is not the possibility of specifying the significance of the specific features in the compositions (Hinkel, 2002). Thus, following the decision to include as many of the linguistic and rhetorical features of text as could be potentially important, categories such as nouns, pronouns, verb tenses and aspects, conjunctions, adjectives, and adverbs were examined. The syntactic and lexical features used in the analysis are classified by their grammatical classes, starting with nouns and pronominals as the first head constituents of sentences (Quirk et al., 1985) and pursued by the verb phrase elements (such as tenses and aspects). The linguistic features are thus categorized into 19 broad types, reported and exemplified in Table 1. Moreover, this study explores the uses of seven textual features that are potentially influential in such rhetorical textual constructs as cohesion, clarification, and hedging (see Table 2).

3.6. Data Analysis Procedures

Comparisons of linguistic and rhetorical features were not accomplished in this study based on averages due to the fact that averages often obscure the spread of frequencies in the sample (Hinkel, 2002). The Kruskal-Wallis test was employed to rank the rates in DI and TI separately for different proficiency learners. Hence, the function of Kruskal-Wallis was the between-group comparison in each instructional condition. For within-group comparisons used to note the differences between the DI and TI among different proficiency groups, a Friedman’s two-way analysis of variance was run.

 

4. Results

Tables1, 2 and 3 present the results of median scores and the significance of the Kruskal-Wallis test comparing the B1, B2 and C1 participants.

 

Table 1.

Linguistic Features Used in DI and TI Contexts

 

 

	DI (Median)					TI (Median)
	B1                             .	B2	C1	Sig.		B1	B2	C1	Sig.
I. Linguistic features 1. Semantic/lexical classes of nouns
Enumerative	6.13	8.00	10.90	.04		4.56	5.66	6.78	.11
Advance/retroactive	3.70	4.19	6.02	.09		2.12	2.36	3.07	.67
Language activity	3.56	3.78	4.16	.69		2.90	3.24	3.57	.44
Illocutionary	2.67	3.02	2.90	.50		1.90	2.26	2.67	.24
Interpretive	3.26	3.10	5.36	.04		2.22	2.60	3.13	.19
Resultative	4.12	4.95	7.51	.03		2.29	3.77	5.26	.05
Vague	1.19	2.02	1.80	.90		1.33	1.50	1.65	.98
2. Personal Pronouns
First-person singular and plural	2.23	2.45	2.89		.78	1.97	2.16	2.09	.70
Second-person singular and plural	2.49	3.09	2.70		.93	2.36	3.11	3.45	.49
Third-person singular and plural	5.56	6.45	6.79		. 88	5.03	5.50	5.98	.95
3. Slot fillers
Nonreferential it	5.69	7.96	8.16		.04	3.30	4.26	4.88	.60
Existential there	5.80	7.12	9.23		.02	3.14	3.89	3.70	.81
4. Indirect pronouns
Universal and negative	3.12	4.25	4.02		.91	2.81	2.90	3.21	.83
Assertive	4.15	4.42	4.70		.94	2.55	2.41	2.72	.89
5. Nominalizations	3.53	3.77	4.10		.77	3.11	3.00	3.79	.75
6. Gerunds	4.02	4.56	4.90		.87	4.11	4.43	4.85	.77
7. Verb tenses
The past tense	4.12	8.09	8.78		.04	3.46	3.64	3.70	.94
The present tense	5.95	7.14	9.20		.03	3.30	4.18	4.32	.51
The future tense	2.83	2.54	2.99		.91	2.26	2.68	3.12	.58
8. Verb aspects
Progressive	4.33	4.58	4.90		.93	2.42	2.23	2.80	.89
Perfect	2.11	2.24	2.67		.89	1.11	1.43	1.25	.92

In general terms, the findings from Table 1reveals that in narrative written texts, C1 learners in the DI condition were more successful in applying the majority of textual features such as ‘interpretive’ (p = .04), resultatives (p = .03), and existential there (p = .02). In some other linguistic categories such as enumerative (p = .04), non-referential (p = .04), past tense (p = .04), and present tense (p = .03), both B2 and C1 learners performed equally well in implementing these features in the written narratives.

In the TI group, however, a completely reverse pattern was found. Only the C2 learners were successful in employing the resultative (p = .05). This finding is expectedly due to the comparatively high level of C1 participants and the characteristic of narrative genre rather than the effect of TI.

 

Table 2.

Linguistic Features Used in DI and TI Contexts (Continued)

	DI (Median)						TI (Median)
	B1	C1	Sig.	B1	B2	C1	Sig.
Private	2.11	3.99	.78	1.40	2.30	2.16	.67
Suasive,	2.50	2.30	.98	2.62	2.40	2.10	.96
Logical/semantic relationships	2.64	3.78	.90	2.00	2.17	2.90	.94
Expecting/tentative	2.70	4.44	.24	2.12	2.20	2.33	.89
10. Modal verbs
Possibility and ability	2.33	6.70	.03	1.77	2.39	3.38	.77
Obligation and necessity	2.10	2.58	.85	1.59	2.05	2.14	.99
Predictive	.90	4.21	.05	.60	1.00	1.02	.96
11. The passive voice	.70	4.49	.009	.00	.00	.99	.10
12. Copula be as the main verb	1.00	4.50	.03	.70	1.10	4.38	.02
13. Infinitives	2.26	4.40	.09	1.78	2.87	2.90	.13
14. Participles as adjectival or adverbial pre- and postpositional forms
Present participles	.90		1.39		5.50	.000	1.00	1.15	2.31	.63
Past participles	.50		.90		1.20	.70	.30	.90	1.00	.90
15. Adjectives
Attributive	2.88	3.19		4.90		.07	2.25	2.40	.10	4.00
Predicative	1.70	2.39		4.10		.09	2.00	2.18	3.60	.12
16. Semantic and lexical classes of adverbs
Time	2.41	3.13		3.90		.67	1.18	1.40	1.80	.43
Frequency	1.00	1.17		2.19		.76	1.23	1.20	1.99	.80
Place	3.01	3.70		5.08		.08	2.87	2.80	3.15	.83
Amplifiers	1.11	1.65		1.90		.95	.08	1.00	1.31	.90
Down toners	.20	.40		.00		.98	.00	.10	.90	.97
Other adverbs	.10	.07		.12		.97	.06	.11	.80	.81

Table 2 reports that the C1 learners in the DI condition were more successful in applying possibility and ability (p = .03), predictive would (p = .05), the passive voice (p = .009), copula be (p = .03), and present participles (p = .0005). Regarding the TI group, only the C2 learners were successful in employing copula be (p = .02). 

 

Table 3.

Features of Subordinate Clauses Used in DI and TI Contexts

	DI (Median)						TI (Median)
	B1	C1	Sig.	B1	B2		C1	Sig.
Private	2.11	3.99	.78	1.40	2.30		2.16	.67
Suasive,	2.50	2.30	.98	2.62	2.40		2.10	.96
Logical/semantic relationships	2.64	3.78	.90	2.00	2.17		2.90	.94
Expecting/tentative	2.70	4.44	.24	2.12	2.20		2.33	.89
10. Modal verbs
Possibility and ability	2.33	6.70	.03	1.77	2.39		3.38	.77
Obligation and necessity	2.10	2.58	.85	1.59	2.05		2.14	.99
Predictive	.90	4.21	.05	.60	1.00		1.02	.96
11. The passive voice	.70	4.49	.009	.00	.00		.99	.10
12. Copula be as the main verb	1.00	4.50	.03	.70	1.10		4.38	.02
13. Infinitives	2.26	4.40	.09	1.78	2.87		2.90	.13
14. Participles as adjectival or adverbial pre- and postpositional forms
Present participles	.90		1.39		5.50	.000	1.00	1.15		2.31	.63
Past participles	.50		.90		1.20	.70	.30	.90		1.00	.90
15. Adjectives
Attributive	2.88	3.19		4.90		.07	2.25		2.40	.10	4.00
Predicative	1.70	2.39		4.10		.09	2.00		2.18	3.60	.12
16. Semantic and lexical classes of adverbs
Time	2.41	3.13		3.90		.67	1.18		1.40	1.80	.43
Frequency	1.00	1.17		2.19		.76	1.23		1.20	1.99	.80
Place	3.01	3.70		5.08		.08	2.87		2.80	3.15	.83
Amplifiers	1.11	1.65		1.90		.95	.08		1.00	1.31	.90
Down toners	.20	.40		.00		.98	.00		.10	.90	.97
Other adverbs	.10	.07		.12		.97	.06		.11	.80	.81

A Mann-Whitney U test was conducted to investigate the between groups differences resulting from the effect of DI and TI with respect to each level of proficiency. Results indicated a significant difference between the DI and TI for B1 (U = 110.16, p = .014), B2 (U = 123.90, p = .009), and C1 (U = 198.04, p = .000) groups. The comparison of effect sizes using Cohen’s d, however, revealed higher effect sizes for C1 (d = 1.58), followed by B2 (d = 1.03) and C1 (d = .98) levels.
According to Table 3, C1 learners in the DI group managed to gain the highest frequency concerning all the subordinate clauses except for the ‘other adverb clauses’ (p = .09) which reached near significance level. In the TI group, the C2 learners effectively used the noun (nominal clauses (p = .04) noun clauses as adjective complements (p = .02), and full adjective clauses (p =.04) features. This result highlights the effectiveness of DI in attracting learners’ attention to the target features particularly B2 and C2 learners who had exposure to these features before and could consolidate their knowledge after exposure to DI. 

 

Table 4.

Linguistic Features Used in DI and TI Contexts

Rhetorical Features	DI (Median)                                                                                                                                  TI (Median)
20. Coordinating and logical conjunctions/prepositions	B1	B2	C1	Sig.	B1	B2	C1	Sig.
Phrase-level coordinating conjunctions	.90	1.01	6.13	.000	.61	.95	3.13	.01
Sentence-level coordinating conjunctions	.21	.25	.90	.70	.23	.29	.33	.92
Logical/semantic conjunctions and prepositions	.30	.60	1.50	.12	.22	.31	.30	.94
21. Exemplification	1.19	2.10	4.11	.04	1.00	1.24	2.15	.90
22. Hedges
Epistemic adjectives and adverbs	.10	.21	.14	.98	.13	.10	.11	.96
Lexical	.72	1.09	1.52	.16	.80	.91	1.28	.91
Possibility	.52	.44	.89	.98	.27	.31	.20	.86
Quality	.00	.45	.90	.14	.00	.03	.00	.90
Hedged performative verbs	.00	.05	.18	.73	.00	.00	.00	1.00

 

Results of Table 4 clarify that in narrative written texts, similar to linguistic features, C1 learners in the DI condition performed better in applying exemplification (p = .04) and phrase conjunctions (p = .0005). In the TI group, however, a significant effect was found only for the phrase conjunctions (p = .01), reflecting the superiority of C1 learners over other level peers.

 

Table 5.

Linguistic Features Used in DI and TI Contexts (Continued)

3. Rhetorical questions and tags	.07	.17	.65	.18	.00	.00	.06	.88
4. Demonstrative pronouns	3.29	3.80	5.77	.05	.90	1.39	1.60	.14
5. Emphatics	.10	.15	.30	.23	.09	.07	.20	.60
6. Presupposition markers )	.00	.09	.30	.54	.00	.04	.05	.71
27. Fixed strings, including phrasal verbs, idioms, and collocations	.00	.00	.05	.90	.00	.00	.00	1.00

 

Results of Table 5demonstrate that in narrative written texts, similar to linguistic features, C1 learners in the DI condition performed greater in employing the demonstrative pronouns (p = .05). In all the other features, there were not significant group differences.

A Mann-Whitney U test was conducted to spot the between groups differences resulting from the effect of DI and TI regarding each level of proficiency. Results referred to a significant difference between the DI and TI for B1 (U = 74.56, p = .02), B2 (U = 90.89, p = .007) and C1 (U = 88.65, p = .000) groups. The comparison of effect sizes using Cohen’s d, however, revealed a higher effect size for C1 (d =2.33), followed by B2 (d = 1.21) and C1 (d =1.08) levels. These large effect sizes in comparison to the linguistic feature effect sizes show that the latter was easier for learners to accomplish compared to the rhetorical features that posed challenges to learners in spite of the differentiated instruction. In other words, the DI was less effective for rhetorical features than linguistic features.

5. Discussion

The findings of this study reveal significant differences between DI and TI groups in their narrations in terms of lexical, and syntactic, clausal and rhetorical features. Participants in the DI group in this study did not avoid the use of these structures that they had not yet fully acquired when presented with the opportunity to match a narrative text to their own linguistic capacities. Hence, it might be presumed that narrative tasks provided similar opportunities for learners to make use of and develop their linguistic competence in writing when coupled with differentiated instruction.

With regard to different areas of lexical competence (Meara, 2002) such as lexical variety and range, DI participants at B2 and C1 levels depicted a significantly different profile from the same level TI participants. This finding adds another piece of evidence for the results of previous research in second language writing and assessment (e.g., Crossley & McNamara, 2012) which conclude that lexical proficiency is a cornerstone of writing competence and FL writers are encouraged to focus on if they wish to enhance their writing skills.

FL narrators in the DI condition were found to display higher levels of using subordinate clause. Even TI learners could show high levels of noun clause usage. These findings seem to be different from the findings of studies on expository texts, in which FL writers were found to write longer clauses (Silva, 1993) but employed fewer subordinate clauses (Hu, Brown, & Brown, 1982; Silva, 1993). The diversities between previous investigations and the present study might be ascribed to the discourse specificities of narratives. Subordinate clauses in general and relative clauses in particular include numerous discourse and linguistic functions in narratives, such as to name, locate, and spot old and new referents in the story, to talk about main characters, to encourage, facilitate, and continue narrative actions, to establish predictions with regard to narrative entities and events, and to summarize past or future events (Kormos, 2011). Thus, it is possibly due to these inevitable functions of subordinate clauses that both learners employed them in their narratives.

Regarding the rhetorical features, the DI learners performed better than the TI learners; however, the effectiveness of DI was limited in comparison to the linguistic features. This finding seems to be in opposition with the results obtained from expository and argumentative genres (e.g., Leki et al., 2008), in which learners were generally found to apply a higher frequency of coordinating conjunctions. The causes for the low frequency of conjunctions in this study might be attributed to the features of the narrative genre. It seems that even the C1 learners participating in this study were not aware of the syntactic and cohesive demands of narratives regarding coordination and conjunction and could not develop the adequate linguistic means to express these relations even after instruction. This finding is also contrasted by that of Kormos (2011) which found the significant performance of FL learners in terms of using cohesive and coordinating conjunctions in their narrative texts.

 

6. Conclusion

The results of the present study make it clear that despite their exposure to writing texts in the course of their study, FL learners have not been able to recognize or did not intend to use other, less superficial features of written texts. Nevertheless, the cause of the learners’ difficulties in using appropriate linguistic and rhetorical devices particularly in the case of learners exposed to traditional teaching seems to be relatively easy to identify: Not only do FL writers have a limited resource for lexis and syntax, but they have not focused on the necessary features of discourse and text because they may simply not be conscious of them. Provided that L2 learners do not know the fact that somewhat harsh norms determine how written discourse and text are produced, they may further be unconscious of the fact that they were expected to learn them (Hinkel, 2002).

Recent studies in applied linguistics have attended to the significance of awareness and noticing in L2 acquisition in different contexts and skills (Khezrlou, 2018; Khezrlou, Ellis & Sadeghi, 2017). Referring to considerable findings from psychology and cognitive science, Schmidt (1995) postulated that on the whole, the importance of noticing, attention, and awareness has not been much in the spotlight in second language acquisition (SLA) research, despite the fact that cognitive scientists have long realized that these mental functions are of utmost importance. Schmidt went on to state that, for instance, in learning L2 writing a high degree of awareness of both linguistic and rhetorical structures can bring about learners’ understanding how written discourse is developed. Schmidt underlined that a fruitful and productive learning environment needs to benefit from all dimensions of language learning such as cultivating explicit skills and implicit opportunities to increase learning. However, awareness alone, according to Schmidt, is not adequate for learning: Knowing how something functions does not lead an individual to actually carry out the task. The teaching of language and its methodologies can take advantage of research findings on how L2 is mastered and what instruction tailored to learners’ needs can contribute to improve the quality of learners’ skills in grammar, vocabulary, and text production. This is in line with Ellis’s (1997) argument that, for example, made a distinction between communicative and formal language uses that can be approached and instructed in distinct ways, based on learning needs. The results of this study clearly corroborated that DI learners who had exposure to the differentiated instruction in accordance with their needs demonstrably performed better in the lexical, syntactic, and text-construction skills in comparison to the TI peers.

It appears puzzling that in many English as a foreign language (EFL) and English for academic purposes (EAP) programs, learners are instructedL2 skills without sufficient attention given to the tasks that they are expected to and will need to perform after finishing their L2 training and move on to carry on with their careers (Leki & Carson, 1997). The results obtained from the present study underscore the necessity of substantial modifications to be made toL2 grammar, vocabulary/reading, and writing curricula in EAPs and other EFL programs that teach learners before their entrance to universities. Nevertheless, as most teachers know from experience, filling blanks in sentences or selecting the accurate choice in multiple-choice tests does little to enhance the quality of L2 writing. A noteworthy cause for the shortfall of L2 grammar teaching is that, in most cases, learners do not see a relation between grammar exercises and text generation. Hence, according to the results of this study, the differentiated instruction was required as a “one size fits all” type of instruction could not meet the different needs of all learners. DI individualized learning as much as possible, implementing scaffolds such as writing frames and acronyms to help second language learners with writing difficulties. By having choice, learners could tap into their interests and their general and cultural background knowledge, which is important as many learners wish to retain their own goals (Fletcher et al., 2009). Teachers also need to consider the significance of feedback by giving grade oriented feedback to learners’ writing tasks(Not Achieved, Achieved, Merit or Excellence), as well as process oriented feedback which is “ongoing and integrated into the entire writing experience” (McKensie & Tomkins, 1984,p. 201-202). In this way, teachers can help their learners close the gap between their real writing capacities and their desired level of performance (Hattie & Timperley, 2007). More importantly, based on the assumption that learners not only have large inter-individual differences in terms of their level of achievement but also demonstrate different growth trajectories over time, Förster, Kawohl, and Souvignier (2018) found that the adjustment of instruction to the individual needs of learners by the selection of suitable and challenging teaching approaches and material can appropriately scaffold high achieving learners and meet their individual skills in a longer term. Nevertheless, since the findings of the present study did not tap into the long term effects of differentiated instruction, more studies are encouraged to further investigate this issue.

Further research might be required to obtain deeper insight into how linguistic and rhetorical features impact the quality of FL writing in different types of tasks and in additional genres. A larger and longer sample of writing would lead to a richer analysis of linguistic relations in texts and for more complex statistical analysis of the data. More qualitative examinations of linguistic and rhetorical markers that investigate variety and correctness of use could also offer more vision into genre and proficiency related variation in FL writing performance.

         

 

Table 3.

Features of Subordinate Clauses Used in DI and TI Contexts

	DI (Median)					TI (Median)
	B1	B2                                     .	C1	Sig		B1	B2	C1	Sig.
Noun (nominal) clauses in the subject or object position, with explicit or omitted subordinators	.80	1.23	3.12	.05		.76	.99	2.75	.04
Noun clauses as adjective complements	1.23	1.50	4.44	.006		1.00	1.12	3.28	.02
Noun clauses as objects of prepositions	.09	.67	1.86	.05		.04	.06	.10	.11
18. Adjective clauses
Full adjective clauses with or without subordinators, including pied-piping adjective clauses	.10	.41	2.60	.03		.06	.07	1.90	.04
Reduced adjective clauses in post nominal positions, postpositional adjectives, and appositives 19. Adverb clauses Full adverb clauses, marked by the subordinators of	.08	.20	1.40	.02		.06	.11	.20	.09
Concession	1.00	1.36	3.90		.000	.07	.09	.49	.48
Condition	.90	1.01	4.14	.000		.12	.24	1.19	.28
Purpose	.06	.58	2.12		.03	.05	.10	1.18	.19
Other adverb clauses	.05	.09	1.18		.09	.00	.00	.00	1.00
Reduced adverb clauses	.10	.39	3.80		.02	.00	.00	.16	.80

مراجع

Calderon, M., Slavin, R., & Sanchez, M. (2011).Effective instruction for English learners. The Future of Children, 21(1), 103-127.

Clandinin, D. J. (2006). Narrative inquiry: A methodology for studying lived experience. Research Studies in Music Education, 27(1), 44-54.

Crossley, S. A., & McNamara, D. S. (2012). Predicting second language writing proficiency: The role of cohesion, readability, and lexical difficulty. Journal of Research in Reading, 35(2), 115-137.

Darling-Hammond, L., & Snyder, J. (2000). Authentic assessment of teaching in context. Teaching and Teacher Education, 16, 523-545.

Ellis, R. (1997). SLA research and language teaching. Oxford: Oxford University Press.

Förster, N., Kawohl, E., & Souvignier, E. (2018). Short- and long-term effects of assessment-based differentiated reading instruction in general education on reading fluency and reading comprehension. Learning and Instruction, 56, 98-109.

Graham, S., & Sandmel, K. (2011). The process writing approach: A meta-analysis. The Journal of Educational Research, 104, 396-407.

Harks, B., Rakoczy, K., Hattie, J., Besser, M., & Klieme, E. (2013). The effects of feedback on achievement, interest and self-evaluation: The role of feedback’s perceived usefulness. Educational Psychology: An International Journal of Experimental Educational Psychology, 1-22.

Hattie, J., & Timperley, H. (2007).The power of feedback. Review of Educational Research, 77, 81–112.

Heaton, J. (1975). Beginning composition through pictures. Harlow: Longman.

Hinkel, E. (2002). Second language writers’ text: Linguistic and rhetorical features. Mahwah, NJ: Lawrence Erlbaum.

Houtveen, A. A. M., Booij, N., de Jong, R., & Van de Grift, W. J. C. M. (1999). Adaptive instruction and pupil achievement. School Effectiveness and School Improvement, 10(2), 172–192.

Hu, Z., Brown, D., & Brown, L. (1982).Some linguistic differences in the written English of Chinese and Australian students. Language Learning and Communication, 1, 39–49.

Huebner, T. A. (2010). Differentiated instruction. Educational Leadership, 67(5), 79-81.

Joseph, S., Thomas, M., Simonette, G., & Ramsook, L. (2013). The impact of differentiated instruction in a teacher education setting: Successes and challenges. International Journal of Higher Education, 2(3), 28-40.

Khezrlou, S. (2018). Form-focused instruction in CALL: What do learners think? RELC, 1-17.

Khezrlou, S. (2019). Task repetition and corrective feedback: The role of feedback types and structure saliency. English Teaching and Learning, 1-21.

Khezrlou, S., Ellis, R., & Sadeghi, K. (2017). Effects of computer-assisted glosses on EFL learners’ vocabulary acquisition and reading comprehension in three learning conditions. System, 65, 104-116.

Kiuhara, S. A., Graham, S., & Hawken, L. S. (2009). Teaching writing to high school students: A national survey. Journal of Educational Psychology, 101(1), 136-160.

Kormos, J. (2011). Task complexity and linguistic and discourse features of narrative writing performance. Journal of Second Language Writing, 20, 148-161.

Leki, I., & Carson, J. (1997). “Completely Different Worlds”: EAP and the writing experiences of ESL students in university courses. TESOL Quarterly, 31(1), 39-70.

Leki, I., Cumming, A., & Silva, T. (2008). A synthesis of research on second language writing. London: Routledge.

Lieblich, A., Tuval-Mashiach, R., & Zilber, T. (1998). Narrative research: Reading, analysis and interpretation. Newbury Park, CA: Sage.

Lipnevich, A. A., & Smith, J. K. (2009). Effects of differential feedback on students’ examination performance. Journal of Experimental Psychology: Applied, 15(4), 319–333.

McCarthy, S. J., & Ro, Y. S. (2011). Approaches to writing instruction. Pedagogies: An International Journal, 6(4), 273 -295.

McKensie, L., & Tomkins, G. E. (1984). Evaluating students’ writing: A process approach. The Journal of Teaching Writing, 3(2), 201-212.

Meara, P. (2002). The rediscovery of vocabulary. Second Language Research, 18, 393–407.

Mendieta, J. (2013). Narrative research: An alternative approach to study language teaching and learning. FOLIOS, 37, 135-147.

Parr, J. M., & Jesson, R. (2012). If students are not succeeding as writing: Teach them to self assess using a rubric. In D. Lapp & B. Moss (Eds.), Teaching that supports engagement and rigorous learning (pp. 174-188). New York, NY: The Guildford Press.

Polkinghorne, D. E. (2007). Validity issues in narrative research. Qualitative Inquiry, 13(4), 471–486.

Pritchard, R. J., & Honeycutt, R. L. (2006). The process approach to writing instruction: Examining its effectiveness. In C. A. MacArthur, S. Graham & J. Fitzgerald (eds.), Handbook of writing research (pp. 275-290). New York, NY: The Guilford Press.

Quirk, R., Greenbaum, S., Leech, G., & Svartvik, J. (1985).A comprehensive grammar of the English language. New York: Longman.

Riessman, C. K. (2008). Narrative methods for the human sciences. Los Angeles, Calif.: Sage Publications.

Sadler, D. R. (1983). Evaluation and the improvement of academic learning. Journal of Higher Education, 54(1), 60-79.

Santangelo, T., & Tomlinson, C. A. (2012). Teacher educators’ perceptions and use of differentiated instruction practices: An exploratory investigation. Action in Teacher Education, 34, 309-327.

Schmidt, R. (1995). Consciousness and foreign language learning; A tutorial on the role of attention and awareness in learning. In R. Schmidt (Ed.), Attention & awareness in foreign language learning (pp. 9–64). Honolulu: University of Hawaii Press.

Silva, T. (1993). Toward an understanding of the distinct nature of L2 writing: The ESL research and its implications. TESOL Quarterly, 27, 657–675.

Smit, R., & Humpert, W. (2012). Differentiated instruction in small schools. Teaching and Teacher Education, 28, 1152-1162.

Sternberg, R. J. (1997). What does it mean to be smart? Educational Leadership, 54(6), 20-24.

Tomlinson, C. A. (2008). The goals of differentiation. Education Leadership, 66(3), 26-30.

Watts-Taffe, S., Laster, B. P., Broach, L., Marinak, B., McDonald-Connor, C., & Walker-Dalhouse, D. (2012). Differentiated instruction: Making informed teacher decisions. The Reading Teacher, 66(4), 303-314.