Japanese high school students' L2 reading motivation
by Takako Nishino (Temple University Japan)
|
|
Keywords: L2 reading motivation, Japanese high school students, Expectancy-value theory, principle component analysis |
Motivation is a force that makes people do (or not do) something (Day and Bamford, 1998).
If a child's reading motivation is high, the amount and breadth of reading increases (Wigfield and Guthrie, 1997),
and if the amount and breadth of reading increases, a child will achieve more academically (Anderson, Wilson, and Fielding, 1988).
Thus, it is possible to say that motivation is one of the main determinants of L1 reading achievement.
In EFL situations as well, learners' motivation to read might have a crucial influence on their reading achievement.
However, little research has been conducted in the field of L2 reading motivation in EFL settings. Thus, there is
a need to examine what sub-components of motivation are most closely related to EFL students' reading achievement.
In this study, I will investigate the construct of Japanese high school students' L2 reading motivation by replicating
Mori's (2002) study on Japanese university students' L2 reading motivation.
In the following sections, I will briefly review the research on L1 and L2 reading motivation.
[ p. 32 ]
| ". . . students' motivation may be, to some extent, domain-specific." |
With regards to L1 motivation, Wigfield and Guthrie (1995) hypothesized that students' motivation may be, to some extent,
domain-specific. For instance, some students may be motivated to read English, but not to listen, speak, or write. Based
on this hypothesis, they developed the Motivation for Reading Questionnaire (hereafter, MRQ) in order to investigate the
construct of reading motivation. In developing the MRQ, Wigfield and Guthrie consulted a number of general motivational
theories in psychology because they thought that research on reading motivation, and especially discussions of reading
engagement, had been based on only some of the components identified by general motivation theorists.
One influential general motivation model that Wigfield and Guthrie (1995) consulted was expectancy-value theory (e.g.,
Eccles, Lord, and Midgley, 1991; Eccles and Wigfield, 1995), which links achievement behavior directly to individuals'
expectancy-related beliefs and task-value beliefs. That is, the perceived probability of success and the value the
individual associates with a specific task are key determinants of motivation. The expectancy-value model is divided into
five components: (a) Expectancy for Success (task difficulty + personal expectation of their ability to do the task),
(b) Attainment Value (the personal importance of doing well on the task), (c) Intrinsic Value (enjoyment + interest),
(d) Extrinsic Utility Value (how the task relates to current and future goals + external rewards), and (e)
Cost (time and emotional expenditures).
While focusing on expectancy-value theory, Wigfield and Guthrie (1995) consulted other theories such as self-efficacy
theory, achievement goal theory, and intrinsic motivation theory, and divided L1 reading motivation into three components
and 11 sub-components (Intrinsic motivation and extrinsic motivation are considered sub-categories of Achievement Values
and Goals). These appear in Table 1.
|
Intrinsic Motivation
|
|
Based on this theoretical framework, Wigfield and Guthrie (1997) examined the relationships between children's reading
motivation and the amount and breadth of their reading. In their study, 4th and 5th grade children (N =105)
completed the MRQ. They found that the children's reading motivation was multidimensional and that intrinsic motivation
was more strongly related to the amount and breadth of reading than extrinsic motivation.
[ p. 33 ]
In the SLA field, Day and Bamford (1998) proposed a L2 reading motivation model, which is also based on expectancy-value
theory. Their model includes four major variables: materials, reading ability, attitudes, and sociocultural environment.
Among them, materials and attitudes are considered to be the primary variables; thus, lack of access to the appropriate
materials or a negative attitude would result in lowered degrees of motivation to read in the L2. Although this model is
designed to apply to L2 reading contexts, it is currently unsupported by empirical evidence.
In a recent study that was conducted in an EFL situation, Mori (2002) developed a reading motivation questionnaire after
consulting Wigfield and Guthrie's L1 reading motivation theory. Three components on the original questionnaire (MRQ),
Competition in Reading, Reading Recognition, and Social Reasons for Reading, were eliminated from her questionnaire because
they are not relevant to university students. Furthermore, Mori's questionnaire included items from Gardner's Integrative
Orientation in his Socio-Educational Model of Language Learning (Gardner, 1985) in order to determine if Integrative
Orientation could be a distinct construct in L2 reading motivation.
Integrative Orientation, one of the factors identified at the first specific level of Gardner's 1985 model, refers to
integrative reasons for language learning: i.e., learners learn a second or foreign language because they want to interact
with target language speakers. At the second, broader level of the model are (a) integrativeness and (b) attitude toward
the learning situation. Integrativeness includes Integrative Orientation, and two attitudinal factors. The third, broadest
level includes (a) effort, (b) desire to learn the language, and (c) attitudes toward learning the language. Gardner (1985)
simplified this model and mentioned that motivation is composed of the three elements at the third level (i.e., effort,
desire, and attitude) and a goal (integrativeness, or more specifically, Integrative Orientation).
Mori (2002) referred to the concept of Integrative Orientation and added seven items to the questionnaire: e.g., learning to read in English is important in that we need to cope with internationalization. She administered the questionnaire to 447 students at a women's university in Japan, and after factor analyzing the results, extracted four sub-components of L2 reading motivation:
Intrinsic Value of Reading in English, Extrinsic Utility Value of Reading in English, Importance of Reading in English, and Reading Efficacy.
Mori concluded that FL reading motivation closely resembles more general forms of motivation described in expectancy-value theory. The finding
has both theoretical and practical value in that it suggests that Integrative Orientation may not be a distinct construct
regarding FL reading motivation, and may not be the key factor to motivate FL learners to read in L2.
| ". . . motivation is a somewhat flexible construct that changes with time . . ." |
However, because no other published studies have focused on this issue, there is a need to confirm Mori's findings by
replicating her study in other EFL contexts. Also, as motivation is a somewhat flexible construct that changes with
time (Dörnyei and Otto, 1998), the construct of younger learners' motivation might be different from that of the university
students. Particularly in Japan, university entrance examinations have a strong influence on high school students'
motivation and goals to study English (Brown, 1995; Taguchi, 2002; 2005), a fact which might suggest probable differences
between Mori's findings and what a study on high school learners might reveal.
The purpose of this study is to investigate the L2 reading motivation of Japanese high school students and compare the
results with those of Mori's (2002) study. To this end, I have posed the following two research questions:[ p. 34 ]
Method
The participants in this study were second year students at a private high school in eastern Japan. Two hundred and
sixty-two students (49% female and 51% male) from six intact co-educational classes participated in the study. In high
school, they took one oral-communication class and four reading classes per week which had extensive explanations of
grammar and translation from English to Japanese.
The school was affiliated with a well-known university, and approximately 90% of its graduates entered that university
without taking the regular entrance examination.
The students were largely high academic achievers according to academic aptitude tests administered by cram schools,
but whether they can enter the department they want largely depends on the grades the students get in their regular courses.
One week before the third semester final exam, a six-point Likert scale questionnaire consisting of 30 items was
administered in Japanese. This questionnaire appears in Appendix 1 and the various items
have been translated in Appendix 2. The thirty questions were the same as Mori's (2002), except for items 13 and 21, which were
revised in order to focus on high school reading classes. Hence, the questionnaire included eight components of the
MRQ and the items that were designed by Mori to tap into the construct of Integrative Orientation.
The internal consistency estimate of reliability for the questionnaire was calculated, and Cronbach's Alpha was
estimated at .67. Considering that reliability tends to be lower if the number of items is small (my questionnaire
included only 30 items), and that a lenient cut-off of .60 is common (Garson, 2006), a Cronbach's Alpha of .67 was
considered acceptable for the present study.
One multivariate outlier was eliminated from the analysis. In addition, in order to run a principal components analysis
(PCA), multivariate normality should be checked. Multivariate normality is the assumption that "each variable and all
linear combinations of variables are normally distributed" (Tabachnick and Fidell, 2001, p. 72). The normality of all 30
variables (representing the 30 items of the questionnaire) was checked with SPSS descriptive statistics. None of the
variables had skewness and kurtosis that were significantly different from zero. This suggests that all the variables
were reasonably normally distributed. Also, multivariate normality suggests linearity among pairs of the variables.
Furthermore, multicollinearity was checked by examining the Pearson product-moment correlation matrix of all items with
each other. The vast majority of those correlations were low. The highest was .67, which is below the .90 that Tabachnick
and Fidell set as the problematic level of collinearity.
As no worrisome violation of the assumptions was found, a PCA was run on the six-point Likert scale questionnaire.
After varimax rotation, a six-factor solution, which accounted for 63% of the total variance, was chosen. The six
factors were extracted using the same criteria applied by Mori: minimum eigenvalues of 1.0 and loadings of more than
.45 on a factor, as revealed in Table 2. [ p. 35 ]
One item (13: "I liked reading classes at junior high school") that did not load on any factors at > .45 was eliminated.
Item means and standard deviations are shown in Table 4 of Appendix 2.
Factor 1 was labeled Intrinsic Value of Reading because seven out of twelve items are sub-components of Intrinsic
Motivation in the MRQ: Reading Curiosity (1, 12, 14, 16), Reading Involvement (15, 29), and Reading Challenge (22).
Two items pertaining to Integrative Motivation (4, 25) loaded on this factor. For high school students the potential
need of studying abroad in the future (4) might be related to the Importance of Reading, and Internet searching (25)
might be related to their enjoyment. Thus, these two items are compatible with the seven items that belong to Intrinsic
Motivation. Two items (8, 30) of Reading Avoidance and one item (28) pertaining to Reading Compliance in MRQ loaded
negatively on Factor 1.
The items loading on Factor 2 concern two different components of motivation: Integrative Orientation (3, 5, 10, 19)
and the Importance of Reading (18, 24, 26). Items 3, 5, 10, and 19 are concerned with internationalization, understanding
lifestyle and culture, getting a job that requires English ability, and learning about world opinions. These four items,
although they are sub-components of Integrative Orientation, are likely to represent the importance of learning to read
in English as an international language. Therefore, this factor was labeled Importance of Reading.
[ p. 36 ]
Factor 3 is labeled Reading Efficacy because all three items (11, 17, 21) are from the Reading Efficacy part of the MRQ.
The mean scores of the responses from items 17 ("English is my weak subject.") and 21 ("My grades are not good.") are quite high
although most of the participants are high academic achiever as mentioned above. It may be that the participants in this
study have tendency to underestimate their own competences when self-reporting.
The items loading on Factor 4 are a mixture of several sub-components: Reading Involvement (2), Importance of Reading (27),
and Reading Avoidance (23). Reading Involvement and Importance of Reading are subcomponents of Intrinsic Motivation in MRQ.
Item 23, which relates to whether the content of a reading material is interesting, loaded negatively on Factor 4. Thus,
this factor may be best considered as a component of the Intrinsic Value of Reading.
Among the three items loading on Factor 5 the item 6 and the items 7 or 9 are negatively correlated with each other. The
negative score on this factor suggests that individuals were taking the class in order to get credit or good grades. Thus,
Factor 3 relates to external rewards and can be defined as Extrinsic Utility Value of Reading.
Although only one item loaded on Factor 6, it suggests a communicative orientation. It can be inferred that this factor
relates to Intrinsic Value (because the students enjoy oral-communication class) or Extrinsic Utility Value (because many
of them wish to travel abroad). However, this point cannot be made clear from only one item, and therefore, it was labeled
as Communicative Orientation.
The present study replicates Mori's (2002) study with respect to the content of the questionnaire, the statistical analysis
procedure, and the theoretical bases. The only difference is the participants. The participants in Mori's study were female
university students, while the participants in this study were co-educational high school students. The factors found in
both studies are listed on Table 3. One major difference is Factor 6 in the present study, which is missing in Mori's results.
[ p. 37 ]
| Research Question 1: | What are the major sub-components of motivation for reading in English for a sample of Japanese high school learners of English? |
For the participants in the present study, six factors were extracted and were categorized into the following five
sub-components of EFL reading motivation: (a) Reading Efficacy, (b) Intrinsic Value of Reading in English, (c)
Extrinsic Utility Value of Reading in English, (d) Importance of Reading in English, and (e) Communicative Orientation.
The first four sub-components correspond respectively to the four constructs of the expectancy-value model: (a)
Expectancy for Success, (b) Intrinsic Value, (c) Utility Value, and (d) Attainment Value.
The results provide evidence that the construct of L2 reading motivation closely resembles the general motivational
structure proposed by expectancy-value theory. This is not surprising, because the main theoretical basis of this study
is Wigfield and Guthrie's L1 reading motivational theory, which is largely based upon expectancy-value theory.
Another reading motivation model that is derived from expectancy-value theory is Day and Bamford's (1998) motivation
model for L2 reading. Although a clear correspondence between the components of their model and the six factors in this
study cannot be found, some relationships are to be noted. As mentioned above, Day and Bamford's model consists of
four major variables which determine the amount of L2 reading motivation: (a) materials (how interesting, linguistic
level, attractiveness, and availability), (b) reading ability in the L2, (c) attitudes toward reading in the L2, and
(d) sociocultural environment including influence and family and friends. Intrinsic Value may correspond to the first
variable, materials: i.e., how interesting, attractive, and relevant they are perceived to be by the students. Reading
Efficacy can be relabeled as reading ability, the second variable, Importance of Reading, relates to the third factor:
attitudes toward reading in the second language. Thus, it appears that to a certain extent, Day and Bamford's model
describes EFL reading motivation. If the questionnaire was revised and based more closely on Day and Bamford's model,
it might show that the constructs proposed in their model provide a broadly accurate description of the reading
motivation of foreign language learners.| Research Question 2: | What differences and similarities can be found between the results of the present study and the results of Mori's study? |
Table 3 illustrates how both studies identify the following four factors: (a) Intrinsic Values, (b) Importance of Reading,
(c) Extrinsic Utility Value, and (d) Reading Efficacy. In this study, Intrinsic Value is equivalent to two factors
(Factor 1 and Factor 4), whereas Mori's study listed only one factor labeled Intrinsic Value. Yet, the results are the
same because Mori's Factor 1 (Intrinsic Value) obtained high loadings from 13 items, whose communality represents both
students' interest in reading in English and their perception of enjoyment. Among the 13 items that loaded on Factor 1 in
Mori's study, 12 items are loading on Factor 1 and Factor 4 in this study. That is to say, Factor 1 and Factor 4 in this
study are almost equivalent to Mori's Factor 1, and thus, we can say that the five main factors found in this study are
similar to the four factors found in Mori's study. It seems that those factors exert a strong influence on EFL learners'
reading achievement. Furthermore, among those factors, Intrinsic Value obtained the highest loadings from the 15 items
and accounted for 41 % of the variance in the present study. This result is compatible with Mori's study in which Intrinsic
Value obtained 35%, the highest proportion among the four factors, findings that suggest that Intrinsic Value, or intrinsic
motivation is one of the critical factors for L2 reading achievement.
One difference between Mori's study and the present study is Factor 6 (Communicative Orientation) that is missing in Mori's.
However, only one item loaded on Factor 6. Thus, it may be possible that the factor relates to Intrinsic Motivation,
Extrinsic Motivation, or the other sub-component if a questionnaire with more items is administered. Further research is
needed to verify this point.
[ p. 38 ]
Another difference can be found by looking at the individual items loading on the factors. Among the items which loaded
Factor 5 (Extrinsic Utility Value) in the present study, the items 7 and 9 (studying for getting good grades, and English
reading as a required course) were eliminated in Mori's study. Instead, Mori's Factor 2 (Extrinsic Utility Value) included
items like 6 and 14 (Curiosity) and 4, 5, and 19 (Integrative Orientation). If we examine the content of these items
closely (i.e., studying abroad, understanding the target culture, taking a non-required reading class, getting a job,
reading newspapers/magazines, and learning about various opinions), we can find a common theme: the participants' belief
that learning to read English is related to future goals. In contrast, the same factor in this study is more strongly
associated with external rewards (getting credits or good grades). Although the labels for both factors are the same,
the features that load on each factor are not. That might have been caused by the difference between the participants in
the two studies. The high school students in this study were taking English reading courses as required subjects and were
required to get good grades, whereas the Japanese university students in Mori's study probably did not care about their
grades as much. In other words, to enter a distinguished department in a university is one of the major goals of the
participants of this study, and that is why they are always sensitive to external rewards such as good grades. Thus, as
predicted, Extrinsic Utility Value emerged as a more significant factor in this study than in Mori's.
| "Intrinsic Value is essential for learning, as empirical findings in psychology and in L1 and L2 research have shown." |
As mentioned earlier, Intrinsic Value seems to be one of the critical factors for L2 reading achievement. Intrinsic Value
is essential for learning, as empirical findings in psychology and in L1 and L2 research have shown. This appears to be
true for at least four reasons. First, according to Maehr (1976), "continuing motivation" (p. 446), which is similar to
intrinsic motivation, prompts individuals to engage in learning activities outside of the school setting. Assuming that
L2 learning is a life-long effort, intrinsic motivation will continue to affect a learner's reading achievement, but
extrinsic motivation will not, once the learner has attained his/her goal. Second, a high level of intrinsic motivation
facilitates a positive emotional experience (Matsumoto and Sanders, 1988). Third, anxiety is not aroused in people who
engage in activities that they enjoy (Schmidt, Boraie, and Kassabgy, 1996). Finally, and most importantly, Wigfield and
Guthrie (1997) found that "children with higher intrinsic motivation read more, and with more breadth, than students
with lower intrinsic motivation" (p. 426). They posited that intrinsic motivation was more strongly related to the amount
and breadth of reading than extrinsic motivation. The findings of this study imply that intrinsic motivation may enlarge
the quantity and breadth of reading in an ESL/EFL setting.
In order to promote learners' intrinsic motivation, the best way might be to provide them with interesting materials.
Hidi and Baird (1986) have pointed out that content interestingness promotes text comprehension and recall. If the students
enjoy reading and immerse themselves in interesting stories, their comprehension will become deeper and their expectancy
for success will grow, which, in turn, could increase their intrinsic motivation. This cycle, which involves a positive
interrelation between competence-related beliefs and values, has been demonstrated empirically (Wigfield, Eccles, Yoon,
Harold, Arbreton, Freeman-Doan, and Blumenfeld, 1997). Therefore, one of the implications of this study is to suggest
that teachers who are teaching English to students similar to the participants in this study, should attempt to find
interesting materials and provide them to their students.
Then the questions are what kinds of reading materials are interesting for the learners and how teachers can provide the
students with such interesting materials. Stone (1994, cited in Day and Bamford, 1998) listed the characteristics of
enjoyable books as follows: [ p. 39 ]
Day and Bamford (2002) offered the following suggestions in their top ten principles for teaching extensive reading:
In sum, it can be said that in order to increase students' intrinsic motivation, teachers should provide a wide variety
of interesting reading materials and let the students, themselves, select one to read.
The present study has investigated the EFL reading motivation of private high school students. The students from this
high school can enter the university provided that they do not get very low grades. This situation is quite unique and
different from those of public high schools, or those of many of the other private high schools in Japan. Also, the major
sub-components of the questionnaire were derived from Wigfield and Guthrie's L1 reading motivation theory. Therefore,
answers to the following questions could be pursued in future research:
| "L2 reading motivation . . . resembles the general motivation constructs proposed by expectancy-value theory." |
The results of the present study demonstrate that foreign language reading motivation is multidimensional, and that it resembles the general motivation constructs proposed by expectancy-value theory. In those respects, the present study obtained the same results as Mori's study. Both studies looked at only a limited number of foreign language learners, but it is my belief that "this line of research is also essential in order to explore the relationship between reading motivation and reading behavior" (Mori, 2002, p. 16).
[ p. 40 ]
Topic Index
Author Index
Page Index
Title Index
Main Index
Topic Index
Author Index
Page Index
Title Index
Main Index
[
p. 41
]
