Contextual Inferencing Strategies and Changes in Reading Attitudes: the Case of Iranian EFL Undergraduates Контекстуальні стратегії умовиведення та зміни в поглядах на читання: на прикладі іранських випускників EFL

Objective. The present study examined the impact of contextual inferencing on Iranian EFL learners’ attitudes towards reading in English. Methods and techniques of the research. To achieve such goals, 60 BA undergraduate students at Islamic Azad University of Shiraz, majoring in English teaching and English


Introduction
Reading is the most demanding skill in learning English. According to F. May (2001), in order to have a full comprehension in reading, the learner requires «a variety of highly fl exible process called comprehension strategies» (p. 119). In many societies, it is important that a person be able to read well so that s/he can be successful in life, but this does not mean that reading leads to success, but it is necessary to be a skillful reader in order to become successful (Grabe, 2009). R. Oxford (1990) claimed that inferencing strategies include using some linguistic or nonlinguistic clues to guess the meaning of the unknown words. However, some researchers stated that context does not always provide enough information and it is possible that learners may make wrong inferences and the strategy of inferencing works well with learners who have the skill of problem-solving (Bensoussan & Laufer, 1984;Hulstijn, 1992: 114). Many studies have been done in order to investigate the eff ect of the contextual inferencing strategies on reading comprehension and some studies have been done to investigate the diff erent aspects of vocabulary and reading.
Over the past few decades, there were much emphasis on learners' attitudes, motivation, beliefs, in second language acquisition (Gardner & MacIntyre, 1993;Wenden, 1998). Masgoret, Bernaus, and Gardner (2001) stated that attitude can infl uence learners' motivation to learn another language. According to T. Merisuo-Storm (2007) and R. Oxford (1990), negative attitudes can retard the process of learning a new language; however, positive attitudes and feelings makes language learning more enjoyable and even more eff ective. It is also defi ned by Smith (as cited in Yamashita, 2004: 33) as «a state of mind, accompanied by feelings and emotions that makes reading more or less probable». According to U. Schiefele, E. Schaff ner, J. Möller, and A. Wigfi eld (2012), reading attitude depends on the experienced feelings and emotions towards reading task which can be either negative or positive. Learners attitudes towards reading has an eff ect on both their motivation and reading success since it aff ects the amount of time a learner spends on reading activity (Lazarus & Callahan, 2000). Moreover, T. Kaniuka (2010) tried to fi nd the relationship between reading strategies and learners' attitudes towards reading. He concluded that those learners who received reading instruction, got higher scores in their attitudes towards reading, and as a result, he concluded that it is better to help EFL learners to have positive attitudes towards reading by using eff ective instructions in reading.
Most teachers agree that it is important to be knowledgeable in vocabulary and they believe that learning a second language demands learning vocabularies. A great number of ideas have been raised in order to prove how important vocabulary knowledge is in language acquisition. N. Schmitt (2004) claimed that there is a high relationship between learning vocabulary and profi ciency in diff erent aspects of language. According to some studies performed by L. Verhoeven and C. Perfetti (2008), reading comprehension has a strong relationship with decoding of words and language skills like vocabulary knowledge. They also claimed that vocabulary knowledge is the most signifi cant factor in reading comprehension. According to Schmitt (2004), one of the most important factors in ascertaining the diffi culty of an English text is the number of unknown vocabularies.

Research Questions
1. Does explicit instruction through contextual inferencing aff ect Iranian EFL undergraduates' attitudes towards English reading?
2. Is there any relationship between Iranian EFL undergraduates' encounters with unknown vocabularies in English texts and their attitudes towards reading in English? J. Walters (2006a-b) performed a research with participation of 44 ESL students at San Diego State University with diff erent nationalities and profi ciency levels. The design of the research was pre-and posttest with the aim of investigating the eff ectiveness of teaching three approaches of using the context to guess the word meanings. The three methods of teaching were one general strategy in order to get the meaning of the unknown words, teaching learners how to recognize and understand the context clues to use, and fi nally practicing with cloze exercises. The participants of the study received six hours of training and then the researcher gave them the post-test. After analyzing the data, the researcher noticed that the participants in experimental group performed better on the post-test and got higher scores in comparison to the control group. The results of the study indicated that training these methods had a positive eff ect on learners' reading comprehension.
Another study done by X. Fuping (2006) to investigate the eff ect of strategy training on learners' reading comprehension. Seventy-six intermediate EFL students participated in the study and were assigned to two groups of experimental and control. The researcher used a learning strategy questionnaire before and after the treatment. A reading test was used as the pre-test, and at the end of the treatment, another reading test were conducted by the researcher as the post-test. The results of the study revealed a signifi cant improvement in the participants' post-test which could be concluded that the lexical strategies had a positive eff ect on learners' reading comprehension

Participants and Setting
Sixty Iranian BA undergraduates majoring in Teaching English as a Foreign Language (TEFL) and Translation Studies at the English Department of Shiraz Azad University were selected through stratifi ed sampling technique. They were homogenized based on Oxford Placement Test. The selected students were randomly assigned to two groups of experimental and control. Both male and female students (12 males, 48 females) within the age range of 19-23 (mean of age: 21) were selected to participate in this study.

Instruments
To collect the necessary data for this study two instruments were utilized. First, an 'Attitudes towards Reading in English' questionnaire was administered to the participants. This questionnaire was developed by Figen Tezdiker (2007) and modifi ed by the researcher. It was composed of four parts: a) personal information about the learners (i.e. name, gender, fi eld of studies, and level), b) 31 items for attitudes of the learners towards reading and for joy of reading, scored on fi ve-point-likert scale (i.e. strongly agree=1, agree=2, undecided=3, disagree=4, strongly disagree=5), c) one general question about their attitudes towards reading in a foreign language, and d) seven items for attitudes of learners towards unknown vocabulary encounters which were added to the questionnaire by the researcher from 'Attitude towards Reading in English' questionnaire by Demet Kulac (2016). The questionnaire was scored on fi ve-point-likert scale to analyze the attitudes of the learners towards reading English texts. F. Tezdiker (2007) has also validated the original version of the questionnaire. Since the questionnaire was modifi ed by the researcher, the validity was confi rmed through pilot study during this research. To do so, thirthy EFL students were selected randomly from the target population, and homogenized through Oxford Placement test. Twenty learners were selected after the Oxford Placement test, and were assigned to two groups of control and experimental randomly. To ensure that the questionnaire used in this study enjoyed reliability, the researcher used the Cronbach Alpha reliability and it was reported as 0.78. In Order to elicit participants` viewpoints regarding the eff ectiveness of contextual inferencing strategies, a revised version of Demet Kulac's interview (2016) was used. There were fi ve main questions for the students to respond to, and the responses were recorded by the researcher while interviewing, and analyzed according to the analysis of the interview used in Demet Kulac's article (2016).

Data Analysis Procedure
In order to test the eff ect of explicit instructions in contextual inferencing strategies and unknown vocabulary encounters on Iranian undergraduates' attitudes towards reading, the data gathered through the questionnaire were analyzed by using the Statistical Package for Social Sciences (SPSS) to see whether strategy training had a signifi cant eff ect on learners' attitudes towards reading in English. In order to fi gure out the eff ect of unknown vocabulary encounters, the correlation between the learner's attitudes towards unknown vocabulary encounters and towards reading attitudes in English was calculated using Pearson Correlation Coeffi cient. As for the interview, qualitative techniques such as eyeballing, coding, sorting, and pattern building were applied to the recorded and transcribed data of the interview according to the qualitative analysis of the interview used in Demet Kulac's research in (2011), and discussed at the end of the study.

RQ. 1. Results of the First Research Question
Paired samples statistics was obtained for learners' pre-and posttest of control and experimental groups and their attitudes in order to fi nd out if explicit instruction through contextual inferencing strategies have any eff ect on Iranian EFL learners' attitudes towards reading English texts as shown in table 1. Thereafter, the paired sample t-test was implemented to analyze and compare the mean scores between learners' pre-test and post-tests, as depicted in table 1.

Paired Samples Statistics of Learners' Pre-tests and Post-tests of Reading Attitudes
As illustrated in table 1, the mean score in pre-and post-tests of control group were M=2.68 and M=2.60 respectively, and the mean scores in pre-and post-test of experimental group were M=2.68 and M=2.38 respectively. Since in the scale of the questionnaire, the strongly agree option considered as 1, and the strongly disagree option got the score 5 (i.e. strongly agree=1, agree=2, undecided=3, disagree=4, strongly disagree=5), the mean score measured in the post-test was lower than the mean score in pre-test; therefore, the lower mean score in the post test showed that the treatment changed learners' attitudes positively. Then, to assess the signifi cance of the diff erence between groups, the researcher run paired samples t-test. The results are presented in table 2.

Paired Samples T-Test of Learners' Pre-Tests and Post-Tests of Reading Attitudes
Based on the Table 2, there was not any signifi cant diff erence between learners' pre-and post-tests of control group (t (29) = 0.938, p>0.05), while there was a statistically signifi cant diff erence in mean scores between learners' pre-test and post-test of experimental group (t (29) = 7.35, p< 0.05), suggesting that learners' attitudes towards English reading could change when they are taught through strategy instructions.
Furthermore, in order to know how big the diff erence between the means of pre-and post-tests of experimental group was, the researcher calculated the eff ect size or the strength of association which showed the relative magnitude of the diff erence. The following formula was utilized to calculate the means diff erence. This formula was used in order to check the eff ect size of the mean diff erences. squared= (^2) / (^2+ ) squared= (7.357^2) / (7.357^2+29) Then, by putting the values in the formula, the obtained eff ect size was (0.65), which means that 65 percent of the variances were eff ective in this study.

RQ. 2. Results of the Second Research Question
The second research question delved into unknown vocabulary to indicate that if there is any relationship between learners' encounters with unknown vocabularies in English and their attitudes toward reading in English. Therefore, the researcher calculated the correlation between the learners' attitudes towards unknown vocabularies in reading texts and towards reading in English. Tables 3 depicted the results, and descriptions related to the results are presented in following. As demonstrated in Table 3, there was a negative relationship between the learners' attitudes towards unknown vocabularies in reading texts and towards reading in English. In fact, the correlation coeffi cient was -.59 and the signifi cant value was less than 0.01. Therefore, it can be concluded that there was a negative relationship between the learners' attitudes towards unknown vocabularies in reading texts and towards reading in English. According to the results obtained, the more learners encounter unknown vocabularies, the lower their attitudes towards reading would be.

RQ. 3. Results of the Third Research Question
In order to know the learners' viewpoints about the eff ectiveness of contextual inferencing, 20 learners were chosen randomly from the experimental group for the semi-structured interview after the treatment. The interview included fi ve main questions for each learner to answer to. The questions were about the learners' ideas and feelings about explicit instruction of contextual inferencing strategies, the usefulness of the inferencing strategies and if they use them in the future, whether the strategy training is able to change the participants' reading attitudes, if the strategy training motivates them to read more in English, and to fi nd out how they felt about the unknown vocabularies in English texts after strategy training. The majority of the participants reported positive feelings about the strategy training. Moreover, they believed that the contextual inferencing strategies are eff ective in reading, and they planned to use these strategies in the future when they want to read a text so that they can infer the meaning of new words in the context. Some of the learners thought that they may have a wrong understanding of the text and the vocabularies, but most of them mentioned the usefulness of reading strategies. The learners mentioned that the contextual inferencing strategies made them more confi dent and gave them a good feeling about themselves when they are able to comprehend the text they are reading. Most of them were motivated to read English texts after learning the reading strategies because they believed using these strategies are fun and enjoyable when they understand the texts, and fi nally, contextual inferencing strategies has helped the majority of the participants in reading comprehension and changed their attitudes towards unknown vocabulary encounters in a positive way, and they were also able to overcome their stress when they encountered new vocabularies in English texts in comparison to the past. Moreover, fi ve major factors were drawn from in participants' responses to the interview questions such as positive attitudes towards reading, the joy of reading, vocabularies in reading, autonomy, and motivation. These factors concluded from learners' responses are presented along with the participants' percentage in table 4.

Conclusion
This study intended to investigate the eff ect of contextual inferencing on Iranian EFL undergraduates' attitudes towards reading skill, the relationship between unknown vocabulary encounters and learners' attitudes towards reading tasks, and also learners' viewpoints regarding the eff ectiveness of using contextual inferencing strategies in their reading task.

Figure 1. Pie chart showing the percentages of major factors
According to the fi ndings of the study, the participants in experimental group performed better in reading tasks and reading attitude questionnaire as their mean scores in the post-questionnaire were signifi cantly diff erent from their mean scores in pre-questionnaire. This diff erence in mean scores showed that the instructions in experimental group help learners to change their attitudes towards reading tasks positively. Therefore, it can be concluded that explicit instructions in contextual inferencing strategies aff ect EFL learners' attitudes towards reading in English.
Moreover, the researcher intended to fi nd out if there is any relationship between Iranian EFL undergraduates' unknown vocabulary encounters and their reading attitude. Based on the results from Pearson Correlation Coeffi cient performed in this study, there was a negative relationship between learners' encounters with unknown vocabularies and their attitudes towards reading skill. In other words, the more learners come across new words, the lower their attitudes will be towards reading skill. The fi ndings of the study indicated that new vocabularies make reading comprehension diffi cult for learners, and these diffi culties lead to negative attitudes towards reading task.
Finally, the researcher held a revised version of a standard interview in order to fi nd out learners' viewpoints about the usefulness of contextual inferencing strategies in reading task. Most of the participants claimed that they found these strategies useful, and they were able to comprehend the text better. According to their responses, using strategies were much better than using dictionary during reading activity, because using dictionary may interrupt the reading process, whereas using contextual inferencing strategies can help them to focus more on the text and infer the meaning of the word by using clues in the context. One or two participants preferred using dictionary because they didn't feel sure about the meaning they guessed from the context by using these strategies. Some other factors were also found in learners' responses due to using contextual inferencing strategies. Leaners felt more independent after learning the strategies, and they did not need their teachers help to complete a reading task. They were also motivated to read more, and they enjoyed reading English text more than they did before as they could comprehend the text better. They were able to guess the meanings of unfamiliar word, and fi nally they had positive attitudes towards reading skill.

Acknowledgments
No project of this size and duration can be completed without a great deal of help along the way. Therefore, I must sincerely thank the following people, all of whom have provided me with either technical advice, critical feedback, emotional support, or fi nancial assistance.
Firstly, I wish to express my deepest gratitude to my supervisor, Dr. Amin Marzban, for his endless guidance, support, patience and encouragement in writing this thesis which helped me to sustain the accomplishment of the research. It's an honor for me to thank him for all he has done.
Then, I extend my heartfelt thanks to My Family for providing me with unfailing support and continuous encouragement throughout my years of study and through the process of researching and writing this thesis. This accomplishment would not have been possible without them.
Finally, I gratefully thank the Learners who participated in this study.