For many students in an EFL context, where actual contact with the target culture is limited, reading can be the major source of input. As many researchers suggest, (e.g., Krashen 1982; Eskey & Grabe 1988), students need to read in substantial quantity not only in order to increase the amount of input but also to develop reading skills and knowledge. However, their seemingly preferred way of reading, namely word-by-word translation, sometimes appears to hinder some learners of English from reading in a desirable manner. Consequently teachers who value extensive reading have to make an attempt to convince their students that it is important and useful to read more.
However, it seems that some students buy 'the sales pitch' whereas others simply do not. It can be assumed that such differences may be somewhat attributed to their motivation or attitudes toward learning English and/or reading in English. As Skehan's (1989) internal cause hypothesis suggests, learners seem to bring a certain quality of motivation to the learning situation. Most educators would intuitively agree with this notion as they might have experience with both highly and less motivated students. My initial interest in investigating EFL motivation was prompted by the following question: what kind of students read more than others?
As both researchers and educators recognize that motivation can be one of the key predictors of success in second/foreign language learning, a large quantity of research has been carried out in order to investigate what constitutes motivation and how it works.
Research in motivation for second language learning had been dominated by the socio-psychologically oriented work of Gardner and his associates until the early 1990s. They established a scientific research procedure for motivational studies by developing a standardized motivation assessment instrument 'the Attitude/Motivation Test Battery (AMTB)', which is still largely used in motivation research. Gardner and Lambert (1959, cited in Crookes & Schmidt 1991) made a distinction between integrative and instrumental motivation using the above mentioned instrument. The integrative motivation associates desire to learn with learners' attraction to the target culture or language whereas instrumental motivation associates their desire with their recognition of more practical merits of learning the target language. Gardner and Lambert (1972) further hypothesized that integrative motivation was more effective than instrumental motivation in promoting language learning success.
Although no one seems to argue against the importance of socio-psychological aspects of motivation in language learning, a number of criticisms have been raised against Gardner's socio-psychological approach to motivation. Together with criticism that the definition of integrative motivation is ambiguous (e.g., Crooks and Schmidt 1991), another criticism is attributed to the fact that studies carried out in different contexts indicated contradictory results. Dornyei (1990), for instance, developed a motivational questionnaire in order to investigate the components of motivation in foreign-language learning, assuming that the parameters created for second-language learning contexts are not directly applicable. The results of the study indicated that integrative motivation was weaker than instrumental motivation for predicting proficiency.
Schmidt, Boraie, and Kassabgy (1996) also argued in their study of Egyptian learners of English that Gardner's model cannot be assumed appropriate to EFL context where learners have limited interaction with the target language.
In Japan where a single language and culture dominates, anecdotal evidence is often reported that some advanced students, especially returnees, purposely downgrade their English due to their reluctance to stand out. Although it would be rash to claim that all returnees have high integrative motivation, this phenomena might suggest that there is a possibility that integrative motivation, especially an inclination towards the target language culture, could have some negative impact on language learning success.
Given the criticism of the Gardner model, the present study looked at motivation referring to expectancy-value theories (Atkinson & Raynor, 1974, Pintrich & Schunk, 1996, Wigfield 1994, cited in Dornyei 1998) instead of instrumental/integrative motivation dichotomy. According to expectancy-value theories, motivation to perform a certain task is strongly related to two components: expectancy of success in a given task and the value the individual associates with success in that task.
Eccles and Wigfield (1995) argued that the value associated with success consists of four components: attainment value (or importance), intrinsic value (interest), extrinsic utility value, and cost. Attainment value refers to the individual's perception of importance of success in a given task. Intrinsic value refers to enjoyment that task engagement brings about, whereas extrinsic utility value refers to the usefulness of the task. Cost refers to extended effort, time and emotional costs.
Another issue addressed in the present study reflects the recent movement toward further conceptualizing second/foreign language learning motivation and proposing multi-dimensional models of motivation. Dornyei (1994, 1996), for instance, divided motivation into three levels : language level, learner level and learning situation level. According to Dornyei, motivation at the language level includes an integrative motivational subsystem and instrumental motivational subsystem, motivation at the learner level includes the need for achievement and self-confidence, and the learning situation level includes course-specific motivational components, teacher-specific motivational components and group-specific motivational components.
Crooks and Schmidt (1991) also attempted to define motivation "in terms of choice, engagement and persistence, as determined by interest, relevance, expectancy, and outcomes" (p.502). Dornyei (1998) portrayed the recent shift in motivational research as follows:
The reform articles expressed an explicit call for a more pragmatic, education-centered approach to motivation research which would be more relevant for classroom application. The main focus shifted from social attitudes to looking at classroom reality, and identifying and analyzing classroom-specific motives (p.125).
As a result of this shift, in recent motivational studies more emphasis is placed on situation- and task-specific motivation. The present study also attempts to investigate the relationship between general English learning motivation, task specific motivation, and behavior on a certain task.
Additionally, one more component was included: reading motivation. Although in the fields of psychology and educational psychology, the issue of the domain specific motivation has been investigated extensively (Wigfield 1997), the concept seems to have been neglected in SL learning. Wigfield, for instance, proposed a multidimensional model of reading motivation. According to Wigfield, reading motivation needs to be conceptualized from three perspectives: competence and efficacy beliefs, achievement values and goals, and social aspects of reading.
Considering the above-mentioned points, the present study attempts to achieve the following goals: (1) to identify the components of English learning motivation for a population of university EFL learners in Japan, (2) to identify the components of English reading motivation for the same population of EFL learners, (3) to identify the components of task-specific motivation for the same population of EFL learners, and (4) to identify the relationships between the components of motivation and the amount of reading.
Method
Participants
The participants (N=52) in this experiment were first year non-English major students at a four year women's university in Japan. All of the students were in intact groups taught by the researcher. Although the participants were separated into two classes, they had all been placed into the intermediate section of a required English reading/listening course based on their placement test. Therefore, the students could be considered comparable at least in terms of their scores on the placement test, which was an adapted version of the TOEIC.
However, in order to ensure that the students were equivalent on the basis of reading ability, a reading proficiency test was administered prior to the onset of the experiment, and then their mean scores were compared using a t-test. According to their scores on the test, they were found to be statistically equivalent and should be deemed as low intermediate readers of English although they were placed in intermediate sections in the institution. Although there were 72 students total in the two classes, omitting those who missed experiment day(s) left the total number of participants at 52.
Materials
Measures. In order to ensure that all students were equivalent in terms of their reading proficiency at the start, a reading section of SLEP test was administered prior to the onset of the experiment.
SRA materials were used to acquire information regarding the amount of reading done by the students. Each student submitted reading reports which consisted of a graph indicating how much they read and a graph of their performance on reading comprehension scores. Then the total number of words each student had read was calculated.
A questionnaire was designed drawing upon some published motivational/attitude scales (Gardner, 1985; Koizumi & Matsuo, 1993; Schmidt et al, 1996). However, the majority of the questions were original. The questionnaire consisted of three parts : part I - questions pertaining to motivation/attitudes toward reading in English (20 items), part II - questions addressing general motivation/attitudes toward learning English (30 items), and part III - questions investigating motivation/attitudes toward the assigned task (24 items). The questionnaire was developed and administered in Japanese (see appendix A and B for the details of the questionnaire.)
The reliability scores of the three components (parts I, II and III) of the questionnaire using Cronbach’s alpha were .89, .82 and .85 respectively.
Procedures
In the class session following the SLEP test, the first two parts of the questionnaire, i.e. reading motivation/attitudes questions, and general motivation/attitudes questions, were administered.
The students were expected to read SRA reading materials throughout the semester which were placed in the library at their leisure. Decisions regarding the amount of reading and reading speed were left up to them although they were all aware that they had to hand in the reading reports which would be part of their grades. In the middle of the semester approximately three months after the students started to read, the participants submitted their reports. A week after they handed in their report, the last part of the questionnaire which contained questions regarding their motivation/attitudes toward the SRA reading materials was administered.
Results
In order to determine the interrelationship among the items included in three parts of the questionnaire, i.e. Parts I (reading motivation), II (general motivation) and III (task-specific motivation), three principal components analyses were performed.
A factor analysis instead of principal components analysis is suggested when there is a particular underlying model of the data in mind. However, principal components analysis was chosen as a procedure for this study for the following reasons:
- the basic principle of expressing two or more variables by a single factor remains the same though the computations become more involved for analysis of multiple variables (STATISTICA 1989).
- it can be used as an adjunct to other multivariate analysis procedures (Dunteman 1989).
The number of factors to be extracted was based on the following criteria:
- Minimum eigenvalues of 1.0
- Each factor contained individual items with a minimum loading of .55
Following the administration of principal components analyses, multiple regression analyses were computed based on factor scores obtained from principal components analyses with the amount of reading as the dependent variable.
Principal Components Analyses
Part I : Motivation/Attitudes toward reading in English. After varimax rotation, a six-factor solution was chosen, which accounted for 76.86% of the total variance in the motivation subset of reading.
In comparing the structural sub-factors of the three principal components of motivation, some similarities and differences should be noted. Although the data need to be interpreted with caution as each part of the questionnaire consisted of a unique set of question, intrinsic value, attainment value and extrinsic utility value appeared as independent factors of two or more principal components of motivation (table 5).
Intrinsic value emerged as an independent factor of all three principal components of motivation (they are labeled “positive intrinsic value”, negative intrinsic value of stories”, “intrinsic value of questions” and “negative intrinsic value) whereas attainment value emerged as an independent factor of the principal components of reading and general motivation (they are labeled “negative attainment value” and “negative attainment value of reading novels”, and “positive attainment value” and “understand novels/movies”).
Items indicative of extrinsic utility value also emerged as two independent factors of both reading motivation and general English learning components. These were labeled, “grade related extrinsic utility value of reading” and “positive extrinsic utility value”, and “residing/traveling overseas” and " grade related extrinsic utility value” of reading negative attainment value of reading” respectively. Especially, the appearance of "grade-related extrinsic utility value” as an independent factor of both components of motivation might reflect the fact that all participants were required to take the English class in which the experiment was carried out, and that the task itself was part of their grades.
Another interesting finding is that all intrinsic, extrinsic and attainment values emerged both in a positive and negative sense. This may indicate that positive and negative intrinsic, extrinsic utility and attainment values are independent constructs. If so, they may need to be treated separately.
Following the identification of the above-mentioned sub-factors of motivation using principal components analyses, a multiple regression analysis was administered in order to investigate the relationship between those sub-factors of motivation and the amount of reading. The results of regression analysis indicated that only seven out of 21 factors found in the study were significant predictors of the amount of reading. Out of all factors to be found significant predictors of the amount of reading, two were the factors indicative of general learning English motivation, three were the factors indicative of reading motivation, and two were the factors indicative of task-specific motivation. This finding suggests that it is necessary to establish a reliable multi-dimensional models of motivation instead of simpler and more general models represented by the instrumental/integrative dichotomy. The seven significant factors were as follows: 1) S’s perceived difficulty in reading; 2) grade related extrinsic utility value of reading; 3) positive intrinsic value of reading; 4) negative attainment value of learning English in relation with understanding novels/movies; 5) grade related extrinsic utility value of learning English; 6) S’s positively perceived cost of going to the library, and 7) S’s preference for tasks in which they can easily see their progress.
In other words, the results of the multiple regression analysis suggest that the more grade-oriented motivation to read students have, and the more they like reading, the more they read. This finding seems to be fairly straightforward especially considering the fact that the task was part of their grade. However, the results of the multiple regression analysis also indicate that the more students see reading as difficult, the less they think it is useful to study English in order to understand novels/movies, the less grade-oriented motivation to study they have, the more they read. Since some of the results seem to be contradictory, it is obvious that further investigation using both qualitative and quantitative techniques is necessary to clarify the results of the study.
The two other task-related factors found to be significant predictors of the amount of reading are also interesting: the less students think it is a pain to go to the library, and the more they like the way the task is structured because they can see their progress easily, the more they read. This finding seems to have important pedagogical implications if one believes that general learning motivation and possibly reading motivation are more of a trait, and thus difficult to change whereas task-specific motivation is more of a state, and thus easier to change. Especially with university learners of English who are likely to have established certain beliefs concerning and attitudes toward learning English in their previous schooling, it is very challenging for the teacher to change such established traits of motivation. On the other hand, it could be easier for the teacher to encourage his/her students to actively engage in a certain task possibly by choosing and/or modifying the task according to students' needs and values. For instance, the results of this study imply that the step-by-step nature of the task which clearly indicated Ss' progress has encouraged the students to read more.
Despite the above-mentioned possible implications, there are some limitations to the present study. One obvious limitation is the sample size. Some statisticians (e.g. Tabachnick and Fidell, 1996) suggest that sample size for factor analyses should be at least 300 in order for correlations to be reliably estimated.
Another limitation concerns attrition. Although there were 72 students total in the two classes, only 52 students attended all classes in which the experiment was carried out. Therefore, if those who were absent from any/all of the classes shared certain characteristics, there is a possibility that the pattern of dropping out itself could be a variable which influenced the results.
Also bear in mind that since all the participants were in intact groups in a single institution and share certain characteristics such as gender, academic background, nationality, first language, and proficiency level, the generalizability of the findings are limited. It will be interesting to see whether different populations engaged in different tasks show similar patterns of correlation between motivational variables and amount of reading. Although the present study completely relied on students’ self-reports, future studies need to utilize a wider variety of qualitative techniques including observation and interviews in order to investigate students’ motivation.
References
- Crookes, G., & Schmidt, R. (1991). Motivation: Reopening the research agenda. Language Learning, 41 (4), 469-512.
- Dornyei, Z. (1990). Conceptualizing motivation in foreign-language learning. Language Learning, 40 (1), 45-78.
- Dornyei, Z. (1994). Motivation and motivating in the foreign language classroom. The Modern Language Journal, 78 (3), 273-284.
- Dornyei, Z. (1996). Moving language learning motivation to a larger platform for theory and practice. In R.L. Oxford (Ed.), Language learning motivation: Pathways to the new century (89-101). Honolulu: University of Hawai'i, Second Language Teaching & Curriculum Center.
- Dornyei, Z. (1998). Motivation in second and foreign language learning. Language Teaching Research, 31, 117-135.
- Dunteman, G. (1989). Principal Components Analysis. Sage university series paper on quantitative applications in the social sciences. Paper 07-069. Thousand Oaks, CA: Sage.
- Eccles, S., & Wigfield, A. (1995). In the mind of the actor: the structure of adolescents’ achievement task values and expectancy-related beliefs. Personality and Social Psychology Bulletin, 21, 215-225.
- Eskey, D., & Grabe, W. (1988). Interactive models for second language reading: Perspectives on instruction. In P. L.Carrell, J. Devine, & D. E. Eskey (Eds), Interactive Approaches to Second Language Reading(pp. 223-238). Cambridge : Cambridge University Press.
- Gardner. R. C. (1985). Social psychology and second language learning. London: Edward Arnold.
- Koizumi, R., & Matsuo, K. (1993). A longitudinal study of attitudes and motivation in learning English among Japanese seventh-grade students. Japanese Psychological Research, 35 (1), 1-11.
- Krashen, S. D. (1982). Principles and practice in second language acquisition. New York : Pergamon.
- Pintrich, L., & Schunk, H. (1996). Motivation in education: Theory, research, and applications. Englewood Cliffs, NJ: Prentice Hall
- Schmidt, R., Boraie, D., & Kassabgy, O. (1996). Foreign language motivation internal structure and external connections. In R.L. Oxford (Ed.), Language learning motivation: Pathways to the new century (pp. 9-700. Honolulu: University of Hawai'i, Second Language Teaching & Curriculum Center.
- Skehan, P. (1989). Individual differences in second-language learning. London: Edward Arnold.
- STATISTICA for the Macintosh, (1994). Volume II. StatSoft, Itc. Tulsa: Oklahoma.
- Tabachnick, G., & Fidell, S. (1996). Using multivariate statistics(3rd ed.) HarperCollins College Publishers. New York: New York.
- Wigfield, A. (1994). Expectancy-value theory of achievement motivation: A developmental perspective. Educational Psychology Review, 6, 49-78.
- Wigfield, A. (1997). Reading motivation : A domain-specific approach to motivation. Educational psychologist, 32 (2), 59-68.