The effect of extensive reading on Algerian university students’ writing performance

This paper investigates the effect of extensive reading on first year Algerian university students’ writing performance. An experiment was designed with two groups. The experimental group had to read 12 stories over three months. The control group received no treatment. Both groups were pre-tested and post-tested, and the subjects’ writing compositions were marked using the TEEP Attribute Writing Scale. The results suggest a positive effect of the Extensive Reading Programme, as the experimental group outscored the control one on the narrative paragraph writing test. The integration of extensive reading into the first year writing syllabus was proposed.


Introduction
The purpose of this study is to investigate the potential effect of extensive reading on first year English students' writing performance.The study consists of two main parts: a theoretical part examining reading-writing connections, extensive reading benefits and a practical one dealing with the research methodology, the presentation and discussion of the results together with the conclusions reached.

1.Theoretical Background
One of the issues underpinning current thinking about writing development is the relationship between reading and learning to write.Many researchers (e.g., Carson Eisterhold et. al., 1990;Shen, 2009;Alkhawaldeh, 2011) emphasized the importance of reading-writing connections, revealing that there is a high correlation between good writers and good readers.As reading and writing researchers (e.g., Langer & Flihan, 2000) attempted to explore these connections, they pointed out that the interdependence of these two language processes implies that reading influences 115 in a foreign language may serve as an incentive motivating students to read more (Rodrigo et. al., 2007).This feeling of accomplishment may promote learner autonomy that fosters a strong sense of learning success.
Reading relevant literature about reading-writing relationships constituted a worthwhile impetus for conducting this study, particularly because it is an area of research that received little attention in a foreign language context (Carson Eisterhold et. al., 1990).Attempting to compensate for the paucity of research seems a crucial goal.The objective of this study is to investigate the effect of ER on writing performance.Contrary to the majority of studies that dealt with the effect of extensive reading on writing stressing mainly language gains, this study seeks to report both the language and attitudinal benefits of extensive reading in relation to writing performance.This study attempts to answer the following research question: Does an Extensive Reading Programme have any effect on first year students' performance in paragraph writing?
The next part is devoted to the Experimental investigation.

Research Methodology
As the study seeks to examine the potential effect of extensive reading upon students' writing performance, the type of classroom research is quasi-experimental.This kind of classroom research involves a quantitative approach to data collection.For the sake of triangulation, it was highly desirable to collect qualitative data because "at least two perspectives are necessary if an accurate picture of a particular phenomenon is to be attained" (Allwright & Baily, 1991, p. 73).A questionnaire seeking to explore students' reactions towards the treatment, i.e., the extensive reading programme (ERP), was designed in addition to two tests (the pre-and post-tests).The study was basically a classroom investigation that sought to compare the performance of two groups on a test.

Population Sampling and Experimental Procedure
The subjects taking part in this study are 18-20-year-old intermediate Algerian students enrolled in the first year English degree course at the University of Algiers II.The subjects had five years of compulsory English at school before entering university.The experimental and control groups had to be matching for subject variables.Just like the 15 experimental subjects, the 15 control subjects were female, average-achievers in the writing pre-test, and they read in English only rarely.
The experimental treatment consisted of asking the experimental students to read 12 stories within a period of 3 months; i.e., they read one story per week.Both groups were tested before and after the set period for the experiment, and the means of both groups on the post-test were compared.

Description of the ERP
ERP is a programme developed by the researcher who first selected 12 stories.The reading materials should meet the criterion of appropriacy in terms of ability level, grading of the reading materials, interest and enjoyment, and variety (Day & Bamford, 1998).I started by offering a strong rationale for engaging the experimental students actively in the ERP, by raising their awareness of the importance of reading.I conveyed personal impressions about the reading materials to the students, and I tried to keep track of students' reactions by devoting time to them to react orally to what they read.Hence, a type of literary circle activity was created.

Data Collection Instruments
This study relies on the following data collection tools: A test (pre-and post) and a questionnaire.The objective of the pre-test (appendix 1) was to have two matching groups in terms of writing proficiency.The post-test (appendix 2) served to gather data in the form of scores which were compared.Regarding the content of both tests, the students were asked to write one-paragraph long composition.The writing prompts of the tests were selected to elicit narrative pattern.
A questionnaire was administered to the experimental subjects (Appendix 3) and another to the control subjects (Appendix 4).The questionnaire administered to the experimental students aimed at providing insights into the experimental subjects' attitudes towards ERP.The questionnaire administered to the control group was meant to gather some useful information for the sake of sampling a group comparable to the experimental one.

Data Analysis Procedure
The scoring procedure for subjects paragraph composing relies on the Test in English for Educational Purposes (TEEP) Attribute Writing Scale (Appendix 5).This analytic marking scheme is used for it favours an explicit set of features or constructs to guide judgments, which is important to reach the aim of objectivity of assessment.The assessment criteria cover both communicative effectiveness and degrees of accuracy.The subjects' writing compositions were marked by an experienced writing teacher.Both groups were scored out of 21 points as the mark of each criterion ranges from 0 to 3. Students' scores were divided into categories: The Low Achievers (those students whose scores ranged from 0→7 points), the Average Achievers (from 7→14 points), the Good Achievers (from 14→21 points).Both groups' scores on the pre-test ranged from 7 points to 14 points, and hence the subjects are labeled as Average-Achievers.

3.Presentation of the Results
A glance through the results will eventually allow us to answer our main research question.

The Results of the Pre-Test
The Experimental and control subjects were pre-tested a week before initiating the ERP.The results appear in Table 1.As it is noticed from the table, the means of both groups were very close, and the standard deviations (SD) were very close too.The SD was small, so the students in both groups were distributed quite equitably.The internal validity was not therefore affected.Any differences between the two groups would be due to the experimental treatment and would not be caused by any initial imbalance between the groups.

The Results of the Post-Test
The scores of the post-test are displayed in Table 2.One meaningful result relates to the measure of variability (SD) for both groups which is again very close.This fact confirms that the groups are balanced.But, the means of the two groups were different.As can be seen from Table 2, the means of both groups increased as both groups had been exposed to the lesson on narration.But, the experimental group outscored the control group.The effect of the experimental intervention on each criterion of the TEEP Attribute Writing Scale is worth considering.The first four criteria are referred to as communicative effectiveness; the other criteria are referred to as accuracy.The results are displayed in Table 3.
It is clear from the table that the experimental students' mean of accuracy increased after the ERP to 1.79.More particularly, the grammar criterion moved to a percentage of 20.51%, becoming thus the criterion the most positively affected by the ERP.There was no major change in the experimental students' mean of communicative effectiveness (+0.07), but it was apparent that the experimental group shifted enormously regarding accuracy (+0.89).
The effect of extensive reading on Algerian university students' writing performance 119

Presenting the Responses to the Questionnaires
The aim of the questionnaire administered in the same week as the post-test was to elicit the experimental subjects' reactions towards the experimental treatment.The completion of the questionnaire took about 30 minutes for the majority of the subjects.The main results related to students' responses to the questionnaire are presented as follows.
-None of the experimental subjects found reading the stories very difficult, and the majority considered the stories "of average difficulty".-The whole sample of the experimental students found the reading materials enjoyable, by justifying that they found the stories exciting to such an extent that they had "flow" experiences, and their imagination was stimulated.-All the experimental subjects found the stories interesting.In their opinion, they enabled them to learn a great deal about life, and they felt that they were immersed in a new culture underlying the language of these stories.-53.3% of the sample stated that the stories were "very useful" in improving their writing and 46.6% stated that reading the stories was "reasonably useful" for their English writing.They believed that the 12 stories were: a means to generate ideas and models of sentences to help them in constructing grammatical sentences, and a valuable tool not only to enrich one's vocabularies, but also to correct the spelling of already known words.It was noted that the majority of the students expressed their concern with formal features of language and their correctness.-The majority of students perceived the ERP as a welcome boost for the development of genuine reading habits.-Students' different reading interests and tastes due to personal preferences should be taken into account in trying to set up an ER library.

Discussion
This section aims at a holistic consideration of the findings.It seeks to answer the RQ.

Students' Writing Performance and Extensive Reading
In contemplating the findings, it is suggested that the ERP has brought about some marked differences of achievement in favour of the experimental group.It seems possible to extrapolate the finding of the study and suggest that reading a large amount of materials in English might develop writing performance in 1st year students.This finding is congruent with the relevant literature, as it goes hand in hand with the widely held belief that in order to be a good writer, a student needs to read a lot.These findings run counter Kirin's (2010) study suggesting that writing abilities did not improve despite additional reading experiences.Hence, we made the key point that greater importance should be devoted to receptive activities (ER) in order not to limit the learning experience to production only which may entail a reduced time available for language contact.Another major finding worth contemplating is the remarkable increase in the experimental subjects' mean of accuracy.The positive effect of the ERP was most apparent in the area of "grammar".It is noted that the numerical results are congruent with the experimental students' responses to the questionnaire, whose majority stated that thanks to reading the stories, their grammar improved, which they found to be, a sign of good writing.The fact that linguistic correctness preceded communicative effectiveness may be justified on the ground that the experimental students prioritized bottom-up reading, and probably less focus was devoted to features of discourse.The transfer of grammatical knowledge and language mechanics from reading to writing came in the first position, because at this stage of language development (students' intermediate level), the reading-writing relationship is primarily based on grammar, spelling and punctuation aspects.At a further stage, knowledge transfer from reading to writing may differ to include other variables; i.e., in upper students' level of language development, the influence of ER on writing may cover other variables like compositional organization.
These assumptions are in line with Shanahan's (1997) Bidirectional Hypothesis which assumes that the reading-writing relationship changes at different stages of language development.In his earlier publication the researcher (1984, p. 467, as cit. in Carson Eisterhold, 1990, p. 92) states "what is learned at one stage of development can be qualitatively different from what is learned at another stage of development".

Students' Attitudes towards the ERP
In considering the extent to which the experimental students engaged in the ERP, we noticed that 66.6% of the experimental sample undertook the reading of more than 10 stories during 12 weeks.Many students engaged in reading the stories though they were not in the habit of reading a lot in English.In terms of the experimental students' evaluation of the experience of reading extensively and its impact on their motivation, 100% of the experimental students viewed the stories as enjoyable and interesting.Hence, it may be assumed that the ERP has impacted the experimental students positively and is thus successful.Another asset of the ERP is that the experimental treatment stimulated the once reluctant readers to read more.This is a bridge to achieve "autonomy" which may create lifelong pleasure readers in English, hopefully proficient writers and life-long learners.

Pedagogical Recommendations
These proposals aim at promoting the teaching of EFL writing to first year students in the Department of English in order to underscore the importance and contribution of some neglected traditional sources of input, like ER, in promoting writing.To this end, setting up an ER library of varied, attractive books at an appropriate language level for students is suggested.The importance of teachers' roles in ERPs should be reinforced as the success of any ERP requires a careful planning and systematic implementation.At length, we propose that an ERP should be an integrated part of a regular first year writing instruction syllabus.

Limitations of the Study
To avoid tentative results, such kind of research studies should be conducted over a longer period of time.But owing to tight schedule and lack of reading materials, the present research could not exceed the 12 week period.Maturation of subjects is a non experimental variable likely to affect the dependent variable.But, in order to validate the results obtained by the experimental subjects, a control group was used.Experimental mortality is an extraneous variable that affected the outcome of the study.A logical solution resided in starting with a large number of subjects, expecting that not all of them were to do the readings on a regular basis.

Conclusion
The objective of the study was to explore the possible effect of the exposure to 12 stories on EFL students' performance in narrative writing.A three-month experiment consisting in urging 15 first year students to read a story on a weekly basis was designed.These experimental subjects were pre-tested and post-tested, and their scores were compared with those of a control group to validate the findings.I attempted to corroborate the quantitative data by asking students to fill out a questionnaire seeking to uncover their attitudes towards the ERP.The results of this investigation seem to provide further supportive empirical evidence that extensive reading affects positively first year students' writing performance.Such findings are to be taken as a positive indication that ER can indeed be beneficially employed as a supplement to first year university English course.

Table 1 :
Scores on the Pre-test

Table 2 :
Scores on the Post-Test

Table 3 :
Compared Assessment Criteria Mean of the Experimental Pre-and Post-Tests