Role of Extensive Reading Habits in Students’ Acquisition of Composition Writing Skills in English in Kenya

Despite the important role reading proficiency plays in learners daily lives both within and outside school, there has been a lot of concern from educationist, researchers and media in Kenya who have reported a saddening trend on poor reading culture among many students in Kenya, This study was prompted by the observed poor reading and writing in English by secondary schools learners in Laikipia County in Kenya. The study therefore sought to establish whether there existed any relationship between extensive reading habits and composition writing amongst learners. The study was guided by Schmidt's (1975) Schema Theory. The descriptive survey research design was used for this study. Fifteen secondary schools in Laikipia County were sampled by proportional stratified sampling technique in five sub-counties across the county. Purposive sampling was used to sample 30 out of 80 (37.5%) English language teachers. Simple random sampling was used to sample 327 students out of about 2670 students (12.24%) in form 2 and 3 in the County. The research instruments used for data collection included: questionnaire for students, interview guides for teachers and composition writing test. The instruments were pilot tested to ensure validity and reliability in a coeducational secondary school in the neighbouring county that was not included in the study. The reliability of the questionnaire and interview guide was estimated using Cronbach’s alpha coefficient which was 0.871 which met the recommended threshold of 0.7 and above. Analysis of data was done using descriptive statistics such as frequency, tables and percentages. Analysis of Variance (ANOVA) was done at 95% confidence level (F(2) = 80.933, P = .001) to establish whether there was statistically significance difference between composition writing mean scores of different categories of school using (SPSS) version 17. Qualitative data were organized according to the study themes and presented descriptively on the basis of the study objective. It was found that learners who engaged actively in extensive reading performed better in composition writing than those who did not. However, it was established that acute shortage of reading resources was a major hindrance to engagement in extensive reading in schools. The study recommended that schools should come up with strong extensive reading policy to promote independent reading so as to improve their writing skills.


BACKGROUND TO THE STUDY
According to Graves and Hanson (2003), writers incorporate what they have learnt about a language, structure and style from the texts they have encountered as readers. They also reflect on knowledge of texts they have and experiences they have had as a way of generating and synthesizing ideas for writing. Providing learners with choices on what they read and write, encouraging them to read and write, providing opportunities to write about topics and ideas that interest them and with which they are familiar, positively affect their attitudes toward writing and offer opportunities to learn (Hanson, 1991). Proficient readers instinctively pick up vocabulary, syntax and spelling of words while reading and they are able to apply them in their writing tasks. Therefore teachers should expose learners to environments with adequate reading materials of their choice so that they can use reading as a means to enhance their writing abilities (Nuttall 1996). Studies which found a positive relationship between reading amounts and writing test scores reported that learners who were exposed to wide range of reading materials and longer duration exposure to the materials had more language input. Consequently, writing abilities of the learners were enhanced. Lai (1993) carried out a study in Hong Kong on primary school learners who read 14.2 grade readers over a period of four weeks. The study revealed that there was improvement in reading comprehension scores but writing scores remained constant and thus they did not show any increase. Hayashi (1999) did a study on extensive reading among university learners in Japan. The learners read outside classroom for 90 minutes per week. They also read on average 759 pages. From their writing scores it was found that they had improved in their writing ability. Mason and Krashen (1997) also found similar results of improved writing abilities when they conducted a study on 104 English major learners reading an average of 1,500 pages of graded readers.
Tudor and Hafiz (1989) initiated an extensive reading programme using graded readers with intention of improving learners' language competency. The results of the study revealed that learners made impressive improvement in reading and writing skills and a simpler but more correct use of syntax in L2. In a one semester extensive reading study among adult lower intermediate learners in the United States of America, Constantino (1995) found that as the reading programme progressed, more and more learners not only wrote journal entries on what they were reading but also the length of these entries increased. Divergent research finding have been reported on the effect of extensive reading on writing abilities. Mostly positive gains have been reported among the ESL learners while negative gains have been reported among EFL learners in both short and long term engagement in extensive reading. For instance, Lee and Krashen (1996) carried out a study on extensive reading on 318 Taiwanese high school learners. It was found that the group had a positive but very modest relationship between measures of free voluntary reading and their writing ability. Pigada and Schmitt (2006) carried out a study to determine whether an extensive reading programme could enhance lexical knowledge. The results of the study revealed that knowledge of 65% of the target words was enhanced in some way for a pick up rate of about 1 of every 1.5 words tested. Spelling of words was greatly improved whereas grammatical knowledge and meaning were also enhanced but not at the same rate. The study strongly indicated that learners acquire more vocabularies through well-organized extensive reading programmes. These findings informed the current study in that for a good extensive reading programme to be successful relevant reading materials should be availed to the learners.
Singh (2011) conducted a study in India on one hundred randomly sampled advanced level secondary students from two high schools to establish the relationship between their academic achievement and study habits based on gender. The findings of the study were that boys and girls differ significantly in the way they study and also academic performance. These findings were of importance to the current study as some of the schools sampled were co-educational. However, the current study did not attempt to look at gender difference in study habits but it looked at extensive reading and how it influences English language performance. Kirigia (1991) studied the reading performance of pupils completing primary education in Meru municipality, Kenya and determined the difficulties accounting for the variations in performance. It was realized that a large number of pupils who took the test did not comprehend messages written in English. This was observed to be because of learners' inability to understand general statements when used in particular contexts, their failure to apply the statements in particular situations, inability to comprehend statements which had unfamiliar vocabulary or vocabulary that had specialized usage, even when they were irrelevant to the understanding of the text and the un-usual text layout including unfamiliar syntax. Kirigia's study, informed this study in that a good number of primary school learners join secondary school having not mastered literacy skills. Unlike Kirigia's study, the current study investigated the influence of extensive reading habits on English language composition writing in secondary. Nganyi, (2006) carried out a research on reading culture in a sample of East Africa secondary schools and found that it was poor. He established that students do not like reading even with the provision of reading materials. He recommended that teaching reading should start at basic grades right through secondary, if the reading culture is to be established effectively. Gathumbi in Groenewegen, (2008) carried out a research which sought to determine the competence in reading of the primary school learners. This was done using a criterion referenced test (CRT) on class 3 and 6 pupils across the country. The study established that many primary class 6 pupils had not attained basic reading abilities. The findings of this study are very important to the current study in that, after these learners take their Kenya Certificate of Primary Examination (KCPE), they go to secondary school where a command in reading competence helps them do better in all the subjects in the curriculum. Yet secondary school language teachers assume that learners joining their schools know how to read. It was, therefore, deemed necessary to examine whether extensive reading habits influenced learners achievement in composition writing.
In addition, Tsang (1996) examined a group of Cantonese speaking students participating in three English programmes which included extensive reading in Hong Kong and found an improvement in writing skills in the extensive reading group whereas Polak and Krashen (1988) conducted three studies to determine whether a relationship exists between spelling competence and voluntary reading for ESL college students in the United States of America (USA) and found that voluntary reading helps improve spelling. The study strongly indicated that learners acquire more vocabulary through well-organized extensive reading programmes and this helps them sharpen their writing skills. Previous findings report both positive and negative relationships depending on the quantity of reading and length of exposure to the language. Nevertheless, inconsistent results taken from just a part of the empirical evidence on what is known about the relationship between reading comprehensible texts and writing competency seem to suggest that there is no systematic correlation between the two variables. This inconsistency calls for this study to scrutinize this relationship further with a view of establishing the extent extensive reading influence composition writing in classrooms.

Purpose of the Study
The purpose of the study was to establish the likely influence of extensive reading habits on English language composition writing in secondary schools in Laikipia County, Kenya.

THEORETICAL AND CONCEPTUAL FRAMEWORK
This study was guided by Social Cognitive Theory. The Social Cognitive Theory which proposes reciprocal determination as a primary factor in both learning and motivation was theorized by a Canadian psychologist Albert Bandura (Bandura, 1997). According to this theory, portions of an individuals' knowledge acquisition can be directly related to observing others within the context of social interaction, experiences and outside media influences. According to one of Bandura's (1997), Social Cognitive Theory construct, observation learning, people can witness and observe a behaviour conducted by others, and then reproduce those actions. This is often exhibited through modeling of behaviour. If individuals see successful demonstration of a behaviour, they can also complete successfully. So learners can develop reading habits through observing others in reading situations. This is supported by Vygotsky (1986) that every function in a young person's cultural development appears at the social and individual level. Therefore, English language teachers should be good role models in motivating learners to read extensively through actively participating in reading sessions in classrooms. It is a worrying trend to establish that very little time is allocated and used on reading instruction in the secondary school despite difficulty in reading subject textbooks being quite common amongst many secondary school learners (Littlefair 1991). It is the responsibility of the task of the English language teacher to ensure that the learners develop effective reading habits, strategies and techniques. This assertion is supported by Benson (2001) who argues that language teachers should not leave the learners to engage in learner autonomy alone but they should actively encourage and provide the necessary support for the learners to enable them to take control of their reading. The conceptual framework in Figure 1 was derived from the theoretical framework.

Figure 1: Relationship between independent, intervening and dependent variables of the study
The framework shows the independent variables in this study as extensive reading and provision of reading materials. These variables were investigated to establish whether they have any influence on learners' achievement in composition writing. The dependent variable was students' achievement in composition writing.
Intervening variables are variables that may influence the independent variables in explaining the outcome of the study. The intervening variables in this study were teachers' factors that include teachers' qualifications, teachers' attitudes, students' attitudes and school environment. To account for these variables, the teachers who participated in the study were those who had a degree in education and were trained to teach reading English in secondary schools. They must have taught for at least two years.

METHODOLOGY Research Design
The study utilized a descriptive survey research design. According to Fraenkel and Wallen (2003) surveys are cost effective and efficient when populations are large and their analyses are direct. The major purpose of this design is to describe the state of affairs as it exists (Kombo & Tromp, (2009). Descriptive surveys are useful for education fact-finding and provide a great deal of information which is accurate. They are concerned with describing, recording, analysing and interpreting conditions that either exist or existed (Serem, Boit and Wanyama, 2013). A descriptive survey design therefore, attempts to describe characteristics of subjects or phenomena, attitudes, opinions, preferences and perceptions of persons of interest to the researcher (Kombo & Tromp, 2009, Mugenda & Mugenda, 2003. The study was supplemented by views of qualitative data in form of respondents' views, comments and opinions on main themes the study had raised. The study also used descriptive statistical methods to analyze and present quantitative data. The study used qualitative method because it involves analysis and presentation of data in forms of descriptive nature (data obtained are expressed in words). For qualitative data, the researcher used semi-structured interviews with English language teachers. Quantitative method of data collection was employed because the study used descriptive statistical methods to present and analyze study data in frequencies and percentages. For quantitative data, the researcher used structured and an open ended questionnaire and a composition writing test with learners. The use of both qualitative and quantitative methods of data collection helped determine the extent of use of pedagogical techniques that promote extensive reading habits and also the possible influence of extensive reading on composition writing.

Target Population and Sample Size
The target population for this study was secondary school students and teachers in Laikipia County. The target population was 2670 students in form 2 and 3, 120 English language teachers and 67 secondary school principals. Form 2 and 3 were chosen because these were classes which had settled down and familiarized themselves with reading materials like class readers. They were also thought to have wider experiences in reading yet they were not examination classes. It is also at form 2 that extensive reading of literary and non-literary text is taught according to KIE (2002). In Laikipia County there are 67 secondary schools; among them are 4 boys' schools, 6 girls' and 57 co-educational schools.

Data Collection Instruments
Data collection instruments were questionnaires, interview guide and composition writing text. The questionnaire was administered to students to gather data on factors that influence learners' extensive reading habits. The questionnaire also gathered data on extensive reading patterns and preferences amongst secondary school learners. In total (99%) 327 questionnaires were responded to with only 3 (1%) registering a non-response. The interview guide was administered to English language teachers to gather data on how extensive reading was conducted in their schools and the challenges encountered during teaching. This information was corroborated with data available in English language teachers' professional documents that they use for teaching such as lesson plans and schemes of work.

Research Results and Discussion
The study established that a majority (93.3%) of the English language teachers believed that English language composition writing was influenced by learners' extensive reading habits. It was found that learners who were well grounded on reading were able to transfer vocabulary, spelling and sentence structures they came across in their reading to their compositions. Learners who did not read extensively struggled to express themselves by making glaring spelling mistakes and poor sentence constructions. Their compositions were jerky and creative use of language was missing. The compositions had numerous direct translations from mother tongue to English. National schools that had relatively adequate reading materials performed better than county and sub-county schools.
Study data on the correlation between learners who were currently reading a book for pleasure and their composition scores was obtained. The learners were asked in the questionnaire whether they were currently reading a book for pleasure. The response to this data was analyzed alongside their individual composition scores. The results are shown in Table 1. Table 1 indicate that there is a statistically significant difference between those currently reading a book for pleasure and their composition scores (t(325) = 3.572, P <0.001). Those who were currently reading a book for pleasure had a higher mean score than those not reading. Those currently reading a book were 1.302 more times likely to score average or above average score than those not reading. C.I (1.068-1.586).

Study data in
From the study data in Table 1, the learners who reported that they were currently reading a book got higher scores than those who did not. Those who were reading a book got a mean score of 22.72 while those who said that they were not reading a book got a mean score of 19.52. This finding is important as it emphasizes the importance of the language teachers initiating extensive reading programmes in order for the learners to sharpen their writing skills. The findings of the study concur with the results of Pigada and Schmitt (2006) who carried out a study to determine whether an extensive reading programme could enhance lexical knowledge. This study assessed a relatively large number of words (9133) and examined whether one month of extensive reading could improve knowledge of these target words spellings, meaning and grammatical characteristics. The study also demonstrated how vocabulary acquisition varied according to frequency in which the words were encountered in the text. The results of the study revealed that knowledge of 65% of the target words was enhanced in some way for a pick up rate of about 1 of every 1.5 words tested. Spelling of words was greatly improved whereas grammatical knowledge and meaning were also enhanced but not at the same rate.
Study data related to book preference were obtained from students' questionnaire. The students were asked what type of books they preferred reading extensively and the data obtained were correlated with data obtained from their composition scores as is shown in Table 2.  Table 2 revealed that students posted different composition writing scores depending on their preferred extensive reading materials. Those students who preferred reading novels and short story were the majority with 78 (23.8%) scoring below average and 175 (53.5%) scoring above average. This preference had a total of 253 (77.4%) learners. A Chi-Square test was generated at 95% confidence level (P value =0.034, df= 7, χ2=15.200) to determine whether the relationship between extensive reading preference and composition scores were statistically significant. The P value was found to be 0.05 confirming that the relationship between reading preference and composition writing scores was statistically significant. It is therefore important for the English language teachers to establish students' reading preferences so as to bolster reading culture in their schools. If this is not done the school may waste a lot of resources purchasing reading resources that do not interest the learners and thus discourage them from reading extensively.
A further one way analysis of variance (ANOVA) was used to analyze data obtained from mean scores of composition writing as a result of reading different reading materials. ANOVA was used because the difference being investigated was between more than two groups. This analysis is shown in Table3.  Table 3 indicate that students who read science fiction, novels and story books had the highest mean score in composition writing. The means were significantly different at 95% confidence level (P value =0.042, df= 7, F=2.108, C.I (0.718-2.480). This was less than 0.05 thus confirming the difference in mean score was statistically significant. Those students who preferred reading novels and story books had a mean score of 11.04 while those who preferred reading science fiction had a mean score of 12.22 in composition writing. Those students who used revision books as materials for extensive reading scored the least with a mean score of 8.19 which clearly indicates the importance of students reading extensively outside their subject areas. It is therefore important to expose students to variety of reading materials so that they can improve their writing skills. There are high chances they will encounter different vocabularies, expressions, creative use of language and different sentence structures in the course of their reading.

CONCLUSION
The study tried to establish the likely influence of extensive reading habits on learners' English language composition writing in secondary schools. The finding of the study was that learners who engaged in appropriate extensive reading habits attained higher composition writing scores than those who did not. It can therefore be concluded that extensive reading engagement improves performance in writing. This suggests that those who read widely are likely to write better than those who do not read.