A Retrospective Survey of L2 Learning

Motivational Changes

Tsuyuki Miura
Temple University Japan

This study investigated the long-term L2 learning motivational changes of 196 Japa-
nese university students using a retrospective approach. The participants’ perceived
level of motivation and the rank order of their motivational reasons over 7 years
were measured using a survey in order to test five a priori hypotheses: (1) The
participants’ motivational levels have frequently changed since they started English
learning; (2) these levels were affected by entrance examinations in their final years
of junior high and high school; (3) the patterns of motivational change between high
and low proficiency university students differ; (4) the rank order of motivational
reasons has changed over time; and (5) the rank orders of motivational reasons
between high and low proficiency university students differ. These hypotheses were
mostly supported. The primary findings indicated that the participants experienced
frequent motivational changes in their learning experiences and they were strongly
influenced by entrance examinations.
本稿は１９６名の日本の大学生を対象に、英語学習における動機付けの変化を調査したも
のである。参加者自らに、過去７年間に渡る動機付けの強さとその理由の順位を回顧的調査手
法にて報告させ、そのデータをもとに、以下５つの仮説を検証した。（１）参加者の動機付けの
強さは、学習開始時から変化し続ける、（２）動機付けの強さは、中３および高３時の入試に強
く影響を受ける、（３）学習到達度の差は、動機付け理由の変化に差を生む、（４）動機付け理由
の順位は時間の経過に伴い入れ替わる、 （５）学習到達度の差は、動機付け理由の順位に差を
生む。すべての仮説はほぼ立証され、結果として、参加者の学習動機付けは頻繁に変化してい
ることや、入試に影響を受けていることなどが明らかになった。

JALT Journal, Vol. 32, No. 1, May 2010

29
30 JALT Journal, 32.1 • May 2010

L
2 acquisition is a long-term process that inevitably involves several
years of study, and as such, many learners and teachers recognize
that learning motivation can change for various reasons over the long
period of time necessary to attain a high level of proficiency. Despite the
importance of gaining a clearer understanding of how and why motivation
for learning a foreign language fluctuates, only a limited number of studies
on long-term motivational change have been reported. One method of inves-
tigating L2 learner motivation change is to use a retrospective approach in
which participants are asked to reflect on the motivational changes that have
occurred in their learning experiences. This study utilized a retrospective
approach in order to investigate Japanese university students’ motivational
changes over a 7-year period.

Studies of Motivational Change

Research into the dynamic nature of L2 motivation can be divided into
two groups according to the length of the learning period investigated. The
first group is made up of longitudinal studies of motivational change that has
occurred over a specific course, which varies in length from 1 to 3 academic
years. The following three studies were conducted by utilizing a primarily
quantitative approach, a common practice in the field of L2 motivation re-
search. Gardner, Masgoret, Tennant, and Mihic (2004) investigated 197 Ca-
nadian students’ motivational changes in learning French over 1 academic
year. They used Gardner’s (1985) Attitude/Motivation Test Battery (AMTB)
in the beginning and end of the 1-year course to measure five affective vari-
ables considered to be important in the socioeducational model of second
language acquisition (SLA): (a) integrativeness, (b) attitudes toward the
learning situation, (c) motivation, (d) language anxiety, and (e) instrumental
orientation. A single-factor repeated-measures ANOVA indicated that there
were statistically significant changes only for a few variables (e.g., French
class anxiety, Motivational intensity, and French teacher evaluation). Gard-
ner et al. also assessed the relationship between the participants’ language
achievement, as measured by the final course grades, and the changes in
the affective variables. The participants were split into three course-grade
groups: A, B, and less than B, and the MANOVA results suggested different
patterns of affective changes in the three grade groups. For example, the A
students started the course with relatively high levels of motivation, posi-
tive attitudes, and low levels of anxiety, and tended to maintain these lev-
els through the year. In contrast, the less than B group had lower levels of
motivation, less positive attitudes towards the course, and higher levels of
Miura 31

French class anxiety than the participants in the other two groups at the be-
ginning of the course, and they became even more negative by the end of the
course. Gardner et al. concluded that the affective changes were moderate
over the 1-year course, the changes were clearly associated with the levels
of students’ success in the course, and there was a clear tendency for the
students’ attitudes, motivation, and anxiety to decrease from the beginning
to the end of the course.
Similar studies have been conducted in the context of Japanese students’
English learning. Irie (2005) tracked 84 junior high school students and
their motivational changes over the 3-year curriculum. The participants’
overall motivation and their motivational profiles were investigated with a
mixed-methods design. The results indicated that most of the participants
maintained a stable degree of L2 motivation over the 3 years, a result that
might have been caused by the students’ use of a number of motivation
maintaining strategies, such as setting proximal and attainable subgoals
and focusing on positive learning experiences. In addition, a supplemental
qualitative investigation revealed that the participants’ teacher was an espe-
cially talented and enthusiastic educator who used a variety of strategies to
increase and maintain her students’ motivation to learn. At the same time,
Irie confirmed that the strength of many students’ motivation decreased
between the beginning and the end of the 3-year curriculum. Irie proposed
two possible reasons for this decrease: the compulsory nature of English
education in Japan and the self-critical nature of some Japanese people.
Another longitudinal study conducted in Japan was an investigation by
Berwick and Ross (1989) into the relationship between the changes in 90
Japanese university students’ motivation and their English learning before
and after they had completed their 1st-year university courses. The re-
searchers used a pre- and posttest design in which a 50-item attitude survey
was administered and the students’ English proficiency was assessed at
the beginning and end of the school year. The survey items were entered as
predictors in a series of stepwise regression analyses that were performed
to identify the best predictors of both the pretest scores and the gain scores
between the pre- and posttests. The results indicated that 150 hours of
classroom instruction resulted in an increase in the number of predictors
from the beginning to the end of the school year (i.e., only three predictors
accounted for 20% of the variance in the beginning of the semester while six
predictors accounted for 43% of the variance at the end of the semester).
The researchers interpreted the emergence of a wider variety of predictors
as an indication that the students’ initial motivational attitudes were tem-
32 JALT Journal, 32.1 • May 2010

poral and that taking the university courses stimulated other motivational
attitudes. Overall, the participants’ motivation was low and there was a
weak relationship between their motivational changes over time and their
performance on the proficiency measures. Berwick and Ross attributed
these results primarily to the university entrance examination system in
which Japanese students’ motivation to learn English peaks in the last year
of high school. Many 1st-year Japanese university students appear to have
little motivation for foreign language learning regardless of whether they
successfully pass the entrance examination for their first choice university,
or (more likely) fail that test and pass the examination given by a university
that was not their first choice.
Distinct from the majority of L2 motivation research, a qualitative ap-
proach to investigating learners’ motivational change was applied in the next
two studies. The researchers focused on individual motivational reasons
expressed by a small number of learners, rather than investigating the over-
all pattern of motivational change in a large group. Ushioda (1998) investi-
gated the characteristics of effective motivational thinking in 20 motivated
Irish college students learning French over more than a year. She conducted
individual interviews twice with a 15- to 16-month interval between the in-
terviews over 3 academic years. The data acquired from the first interview
revealed that the most successful students perceived their positive learn-
ing experiences, such as being in France or a Francophone country, as the
main factors underlying their motivation, and the less successful students
tended to perceive their future goals as the main motivators. Based on the
results, Ushioda concluded that effective motivational thinking is a selective
thinking pattern in which some participants filter their learning experiences
by foregrounding positive experiences and deemphasizing negative experi-
ences; this strategy appeared to help the more successful learners to sustain
long-term involvement in L2 learning.
As a part of his mixed-methods dissertation study, Nakata (2006) also in-
vestigated qualitatively how 1-year student-centered learning experiences
affected the developmental process of motivation of Japanese non-English
major freshmen. The researcher investigated motivation using a social con-
structivist framework. He emphasized the importance of learners develop-
ing a core level of intrinsic motivation and becoming autonomous learners
in order to attain high levels of proficiency. After this 1-year project-based
course was completed, Nakata conducted case studies of five of the success-
ful course participants. He concluded that all five students had developed
intrinsic motivation and that two of them had further developed the core
Miura 33

level of intrinsic motivation necessary for further linguistic development.

He concluded that language learners’ motivation is strongly influenced by
their learning experiences and by the way and degree to which they inter-
nalize what they have experienced.
The second group of studies was focused on motivational changes over
extended periods of time. As the researchers employed a retrospective
approach, these studies are highly relevant to the present study. Hayashi
(2005) investigated patterns of motivational change among 481 Japanese
college students over 9 years: 3 years in junior high school, 3 years in high
school, and 3 years in university. He explained how these patterns emerged
using the framework of self determination theory (Deci & Ryan, 1985). The
participants responded to questionnaire items asking about their L2 motiva-
tion during the 9-year period, specifically the periods when their motivation
was the strongest and the weakest, and the reasons why it was strong and
weak during those periods. Hayashi used cluster analysis to identify four
motivation developmental patterns: high-high, low-low, high-low, and low-
high. The participants displaying the high-high pattern showed consistently
high motivation, while the low-low participants reported having low moti-
vation throughout the 9 years. The high-low pattern was distinguished by an
initially high level of motivation that dropped by the 2nd year of high school,
while the low-high pattern indicated low initial motivation that increased
around the 1st year of high school. Hayashi tentatively proposed that dif-
ferent levels of internalization of extrinsic motivation caused the different
patterns. He argued that initial motivation was the result of intrinsic mo-
tivation, and that initial motivation could be sustained only if the students
internalized extrinsic reasons (e.g., succeeding on an entrance examination)
for studying English. Although he did not statistically analyze the overall
pattern of the 9-year change in the participants’ motivation, a line graph that
he provided showed that the participants’ motivation declined moderately
from junior high school to university, increased slightly when the students
were in their final years of junior high and high school, and declined rela-
tively sharply after entering the university.
Sawyer (2007) investigated the motivational fluctuations of Japanese
learners over 8 years of English instruction (i.e., 3 years in junior high school,
3 years in high school, and 2 years in university) with 120 non-English ma-
jors in a private Japanese university. Sawyer created an instrument in which
the participants were asked to mark their levels of motivation to learn Eng-
lish at the beginning and end of each year in school. The participants also
wrote comments concerning their learning and learning motivation. The
34 JALT Journal, 32.1 • May 2010

statistical analyses supported three previous findings: (a) motivation is high

at the onset of junior high school but decreases, (b) motivation decreases
from the 1st to 2nd year in high school but increases in the 3rd year, and
(c) motivation is high immediately before the university entrance exams
but decreases upon entry into a university. In addition, the hypothesis that
teachers influence students’ motivation gained a number of supportive com-
ments in the junior high school period, while the hypothesis that motivation
is influenced by peers and social group members was more salient in the
high school period.

Motivation-Related Perspectives
In this section, three motivation-related perspectives that constitute the
basis of this study will be briefly reviewed. These perspectives will then be
applied to the motivational rank order section of the research instrument.
The first motivational perspective concerns intrinsic and extrinsic motiva-
tion, which are important components of self-determination theory (Deci
& Ryan, 1985), one of the most influential theories in motivation research.
According to the theory, Intrinsic Motivation (IM) and Extrinsic Motivation
(EM) are distinguished according to the degree of the learner’s self-determi-
nation. Intrinsic motivation, which is considered to be a relatively strongly
self-determined form of motivation, refers to motivation that is based on
internal factors, such as enjoyment or satisfaction. In contrast, extrinsic
motivation refers to motivation that is based on external factors, such as
getting good grades or tangible rewards. EM is considered to be a relatively
weakly self-determined form of motivation. Recent researchers have dis-
cussed several subtypes of intrinsic and extrinsic motivation in accordance
with different degrees of self-determined forms of both. Proponents of this
approach have proposed that extrinsic motivation becomes increasingly
similar to intrinsic motivation as the degree of self-determination increases.
For example, when a learner studies a foreign language because of future
career goals (i.e., for extrinsic motivation) and is aware of the fact that the
decision to study is made by herself for her own sake, her motivation may
be internalized, resulting in a type of motivation that shows no major differ-
ences from intrinsic motivation.
In addition to internalized forms of motivation, goals are also considered
to play an important role in motivated behavior. The importance of goals is
best explained by goal-setting theory (Locke & Latham, 1994), which is the
proposal that goals are necessary for individuals to take action; therefore,
motivation is more likely to emerge when a goal is present. Learning per-
Miura 35

formances are differentiated according to the degree of goal specificity, the

difficulty of attaining the goal, and the individual’s commitment to achieving
the goal. The more specific and difficult that a goal is, the higher the achieve-
ment and the greater the commitment to the goal that people will make,
provided that the goal is perceived as valuable and attainable. Goal-orien-
tation theory (Pintrich & Schunk, 1996) describes the distinction between
intrinsically oriented goals (mastery orientation) and extrinsically oriented
goals (performance orientation), concepts that are related to the distinction
between IM and EM. The differences between intrinsic or mastery orienta-
tion and extrinsic or performance orientation do not necessarily mean that
the former results in greater learning because these two goals can positively
interact and facilitate motivation and learning (Hidi & Harackiewicz, 2000).
The last form of motivation that is pertinent to this study is social in na-
ture, as it arises from the influence of significant others (Urdan & Maehr,
1994). Social motivation includes social welfare goals, social solidarity
goals, and social approval goals. These social motives pertain to the reasons
why students are trying to achieve a goal, rather than what they are trying to
achieve. Wentzel (1999) stated that interpersonal relationships and sociali-
zation processes, such as peer interactions, influence student motivation,
and that the goals that emerge from these social interactions influence the
quality rather than the amount of motivation. In foreign language learning,
students’ parents, teachers, peer groups, and the school environment may
function as the four most important social influences in the learning envi-
ronment (Dörnyei, 2001). However, social motivation and influences from
significant others are subject to cultural contexts. For example, Japanese
students are generally described as more interdependent than American
students (Markus & Kitayama, 1991), and if this is true, they are likely to
be influenced relatively strongly by family members, teachers, and friends.

Research Purpose and Hypotheses

The primary purpose of this study is to investigate motivational changes
that Japanese college students have experienced as they moved through
secondary school to their 1st year of university education. Specifically, I will
test five hypotheses. Hypotheses 1 through 3 concern changes in the partici-
pants’ motivational levels:
Hypothesis 1. The participants’ motivational levels have frequently changed
since they started learning English.
Based on investigations of motivational changes over extended periods
36 JALT Journal, 32.1 • May 2010

of time (see Hayashi, 2005; Sawyer, 2007), I hypothesize that the Japanese
participants in this study have frequently experienced motivational fluctua-
tions.
Hypothesis 2. The participants’ motivational levels were affected by en-
trance examinations in their final years of junior high school and high school.
Hayashi (2005) and Berwick and Ross (1989) found empirical results
indicating the powerful influence that entrance examinations can exert on
Japanese students. I hypothesize that the participants’ motivation increases
before they take the examinations and decreases after the examinations are
completed.
Hypothesis 3. The patterns of motivational change between the high and
low proficiency university students differ; the high proficiency students
have maintained generally higher levels of motivation in secondary school
than the low proficiency students have.
This hypothesis is based on the assumption that the participants’ current
level of English proficiency reflects their past motivational levels.
Hypotheses 4 and 5 concern the change in the rank order of motivational
reasons:
Hypothesis 4. The rank orders of motivational reasons have changed over time.
The participants in this study attended at least three schools where they
were taught by different teachers and where they studied with different
classmates over the 7-year period under examination. They also took two
entrance examinations when they were in their final year of junior and
senior high school. Even though English was compulsory for most of the stu-
dents, these experiences may have influenced their reasons to learn English.
Hypothesis 5. The rank orders of motivational reasons between the high
and low proficiency university students differ.
As Hayashi (2005) reported that students who maintained a high level of
motivation were both intrinsically motivated and had sufficiently internal-
ized external reasons for studying, I assume that these motivational differ-
ences are related to the students’ current proficiency (see Nakata, 2006, for
related implications).

Methods
Participants
Participants in this study were 196 non-English majors studying in a pri-
vate university in western Japan: 161 freshmen, 28 sophomores, 5 juniors,
Miura 37

and 2 seniors.1 All of the students had completed 6 years of compulsory Eng-
lish education in junior high school and high school before entering univer-
sity. The freshmen and sophomores were taking 6 hours of English classes
per week in an academic English program that runs for four consecutive
semesters. Because the university department has a reputation for having a
demanding English program, these participants’ overall level of motivation
to study English may have been higher than that of the average Japanese
university student.
The students took an institutional TOEFL (Test of English as a Foreign
Language) when beginning the program in April 2005, and were placed
into one of two proficiency levels based on the results. The mean (M) of the
TOEFL scores (N = 194 because of two missing cases) was 436 (SD = 42.42).
The TOEFL scores were used in this study to divide the participants into
two proficiency groups. The scores of the high proficiency group (N = 111)
ranged from 437 to 523, and the scores of the low proficiency group (N = 83)
ranged from 330 to 433.2

Instruments
The Appendix shows an English translation of the Changes of Learner
Motivation Questionnaire, in which a retrospective approach was employed.
Part I of the questionnaire asked about demographic information. Part II
presented a motivation chart that was designed to allow the participants to
more readily recollect and graph their past L2 learning motivational levels.
The participants were asked to draw their motivational levels on the chart
for a 7-year period: 3 years in junior high school (JH), 3 years in high school
(HS), and 1 year in university (U1). The x-axis represents the seven school
grades and the y-axis represents motivational level. The y-axis scale has five
levels (i.e., three primary scales for low, mid, and high motivational levels,
and two intermediate levels that are located between the low and middle,
and the middle and high motivational levels). The seven motivational levels
measured with this scale formed a set of dependent variables that was used
to test Hypotheses 1, 2, and 3.
Part III of the questionnaire was a motivational ranking task. In the three
subsections, the participants were asked to reflect on their overall learning
motivations when they were in junior high school, high school, and the 1st
year of university, and to rank order in importance the six statements from
1 (i.e., the strongest motivation) to 6 (i.e., the weakest motivation). The six
motivational reasons that were listed in each subsection were underpinned
38 JALT Journal, 32.1 • May 2010

by three motivational perspectives: the intrinsic-extrinsic distinction, goal

theories, and social motivation theory, as described earlier.
Hypotheses 4 and 5 were tested with data gathered from Part III of the
questionnaire. Two intrinsic motivational reasons and three extrinsic mo-
tivational reasons were included among the six dependent variables in the
ranking instrument. The first and second intrinsic motivational reasons
represented interest toward the target language and culture, and enjoyment
of learning English, respectively. Of the three extrinsic motivational reasons,
the first represented a short-term goal, the second represented medium-
term and specific goals commonly observed in the Japanese context, and the
third represented relatively long-term goals. The last motivational reason in
the ranking instrument represented the influence from significant others.
If the participants perceived that different motivational factors were par-
ticularly memorable at a certain stage of learning, they were asked to write
them in the relevant section.

Procedure
Four instructors teaching in the English language program, including the
researcher, administered the questionnaire during class time in January
2006 on the last day of the fall semester. The instructors told the partici-
pants that the purpose of the questionnaire was academic research and that
their responses were confidential and would not affect their grades. The
instructors also obtained the students’ verbal permission to use their most
recent TOEFL scores.

Data analysis
The data obtained from the survey were initially entered into Microsoft
Excel and then exported to SPSS for statistical analyses. The motivational
levels recorded on the motivation chart were transformed into numbers
from 0 to 5 (low = 0; high = 5).

Results
In the results and discussion sections, the entire sample is referred to as
All Students, and the higher proficiency group and the lower proficiency
group are referred to as the High Group and the Low Group, respectively. In
addition, abbreviations will be used for expressing school years, (i.e., JH = 3
years of junior high school, JH1 = the 1st year of junior high school, JH2 = the
Miura 39

2nd year of junior high school, HS = 3 years of high school, HS1 = the 1st year
of high school, HS2 = the 2nd year of high school, HS3 = the 3rd year of high
school, U1 = the 1st year of university).
Hypotheses 1, 2, and 3 were tested by conducting a series of repeated-
measures ANOVAs with year in school as the independent variable and the
7-year motivational level estimates as dependent variables. The TOEFL
score was used as a grouping variable.

Hypothesis 1: The participants’ motivational levels have frequently changed

since they started learning English. Table 1 presents the descriptive sta-
tistics for the participants’ motivational level as measured on Part II of the
questionnaire. Two tendencies can be generalized. First, a general trend was
identified for All Students, and the High and Low Groups’ motivational lev-
els: They rise steadily in the first 3 years of study, drop in HS1, rise steadily
again in HS2 and HS3, and drop once again in U1. Second, the High Group
displayed a higher motivational level than the Low Group throughout the 7
years. The confidence intervals of the two groups overlap in JH, indicating
that the motivational levels of the High and Low Groups do not largely differ
during that period; however, in HS and U1 the confidence intervals do not
overlap, which indicates that the motivational levels of the High and Low
Groups are reliably different during these periods.

Table 1. Descriptive Statistics for Motivational Levels

JH1 JH2 JH3 HS1 HS2 HS3 U1
All M 3.01 3.13 3.45 3.21 3.38 3.85 3.77
Students 95% CI Low 2.80 2.94 3.25 3.03 3.19 3.66 3.60
Upper 3.21 3.32 3.64 3.40 3.57 4.03 3.95
SD 1.42 1.35 1.38 1.31 1.37 1.28 1.22
Skewness .01 -.17 -.40 -.22 -.27 -.78 -.75
SE Skewness .18 .18 .18 .18 .18 .18 .18
Kurtosis -1.27 -1.18 -1.11 -1.00 -1.12 -.60 -.42
SE Kurtosis .35 .35 .35 .35 .35 .35 .35
40 JALT Journal, 32.1 • May 2010

JH1 JH2 JH3 HS1 HS2 HS3 U1

High M 3.10 3.14 3.61 3.45 3.72 4.13 4.10
Group 95% CI Low 2.75 2.88 3.36 3.20 3.46 4.35 3.88
Upper 3.28 3.39 3.87 3.69 3.97 4.25 4.32
SD 1.39 1.37 1.35 1.31 1.35 1.19 1.17
Skewness .08 -.17 -.57 -.47 -.72 -1.32 -1.33
SE Skewness .23 .23 .23 ..23 .23 .23 .23
Kurtosis -1.22 -1.23 -.94 -.80 -.63 .77 .92
SE Kurtosis .46 .46 .46 .46 .46 .46 .46
Low M 2.99 3.12 3.22 2.90 2.93 3.48 3.34
Group 95% CI Low 2.67 2.83 2.92 2.63 2.65 3.19 3.09
Upper 3.31 3.41 3.53 3.17 3.21 3.77 3.60
SD 1.47 1.34 1.38 1.24 1.28 1.32 1.17
Skewness -.07 -.17 -.21 .04 .24 -.24 -.24
SE Skewness .26 .26 .26 .26 .26 .26 .26
Kurtosis -1.35 -1.10 -1.21 -.95 -.86 -1.16 -.69
SE Kurtosis .52 .52 .52 .52 .52 .52 .52
Note: School years are: JH1 = the 1st year of junior high school, JH2 = the 2nd year
of junior high school, JH3 = the 3rd year of junior high school, HS1 = the 1st year of
high school, HS2 = the 2nd year of high school, HS3 = the 3rd year of high school, U1
= the 1st year of university.

Figure 1 illustrates how the participants’ perceived motivational levels

changed throughout the 7-year period.

Figure 1. Change of Motivational Levels

Miura 41

Note. School years are: JH1 = the 1st year of junior high school, JH2 = the 2nd year
of junior high school, JH3 = the 3rd year of junior high school, HS1 = the 1st year of
high school, HS2 = the 2nd year of high school, HS3 = the 3rd year of high school, U1
= the 1st year of university.

A repeated-measures ANOVA was run with year in school as the inde-

pendent variable, the estimated motivational level as the dependent vari-
able, and the students’ TOEFL score as the grouping variable. The results
indicated that the motivational levels of All Students, the High Group, and
the Low Group changed to a statistically significant degree over the 7-year
period: Wilks’s Λ = .72, F(6, 188) = 12.09, p < .01, multivariate η2 = .28 for All
Students; Wilks’s Λ = .59, F(6, 105) = 12.06, p < .01, multivariate η2 = .41 for
the High Group; and Wilks’s Λ= .80, F(6, 77) = 3.24, p < .01, multivariate η2 =
.20 for the Low Group. Follow-up polynomial contrasts indicated significant
linear effects with means generally increasing over time for All Students
and the High Group: F(1, 94) = 32.50, p < .01, η2 = .14 for All Students, and
F(1, 113) = 47.41, p < .01, η2 = .30 for the High Group. Neither a significant
linear effect nor higher-order effects were found for the Low Group. Table
2 presents the results of post hoc pair-wise comparisons for each group.
Five pairs differed to a statistically significant degree: three pairs in the High
Group, and two pairs in the Low Group. Taken together, these results suggest
that the participants’ perceived motivation has frequently changed since
they started studying English, a finding that supports Hypothesis 1.

Table 2. Posthoc Pair-Wise Comparisons

All Students High Group Low Group
Pair t d t d t d
JH1 – JH2 -1.50 -.11 -1.09 -.10 -1.04 -.11
JH2 – JH3 -4.03* -.29 -4.21* -.40 -1.02 -.11
JH3 – HS1 2.58* .18 1.42 .14 2.26* .25
HS1 – HS2 -2.34* -.17 -2.76* -.26 -.26 -.03
HS2 – HS3 -5.13* -.37 -3.39* -.32 -3.91* -.43
HS3 – U1 .78 .06 .23 .02 .91 .10
Note. School years are as in Table 1 above.
p* < .05.
42 JALT Journal, 32.1 • May 2010

Hypothesis 2: The participants’ motivational levels were affected by en-

trance examinations in their final years of junior high school and high school.
As shown in Table 1 and Figure 1, the means for both the High and Low
Groups declined slightly twice, between JH3 and HS1 and between HS3 and
U1. As shown in Table 2, the pair-wise mean comparisons differed signifi-
cantly in the All Students and the High and Low Groups between JH2 and
JH3, and between HS2 and HS3: All students, t(194) = -4.03, p < .05, d = -.29;
High group, t(111) = -4.21, p < .05, d = -.40 between JH2 and JH3, and; All
Students, t(193) = -5.13, p < .05, d = -.37; the High Group, t(110) = -3.39, p <
.05, d = -.32; the Low Group, t(83) = -3.91, p < .05, d = -.43 between HS2 and
HS3. The estimated effect sizes for these differences were relatively large,
except the one for the Low Group between JH2 and JH3 (d = -.11). These re-
sults indicate that the students’ motivational levels increased between JH2
and JH3, and between HS2 and HS3, and decreased between JH3 and HS1,
and between and HS3 and U1 to a statistically significant degree. Because
these increases and decreases occurred at the same time that the entrance
examinations took place, it is highly likely that these motivational changes
and the tests were related to each other; therefore, the second hypothesis
was supported.

Hypothesis 3: The patterns of motivational change between the high and

low proficiency university students differ; the high proficiency students
have maintained generally higher levels of motivation in secondary school
than the low proficiency students have.
This hypothesis concerns the difference between the motivational changes
that took place in the two proficiency groups. As shown in Figure 1, the means
of the two proficiency groups were similar in JH1 and JH2, but started to differ
in JH3, and the distance between the two groups was maintained for the next
4 years. In addition, the mean increases between JH2 and JH3, and between
HS2 and HS3 were all larger in the High Group than in the Low Group, while
the mean decreases between JH3 and HS1, and between HS3 and U1 were all
larger in the Low Group than in the High Group. As shown in Table 1, the par-
ticipants in the High Group perceived their motivational levels as being higher
than did the participants in the Low Group across all 7 years. This is one in-
dication of the existence of a positive relationship between motivational level
and general proficiency. These findings support the third hypothesis: that the
two proficiency groups in university differ in motivational change and that the
high proficiency students maintained a generally higher level of motivation in
secondary school than the low proficiency students.
Miura 43

Hypothesis 4: The rank orders of motivational reasons have changed over

time.
Hypothesis 4 was examined by conducting a series of Freidman’s tests
with each of the six motivational reasons in each period of schooling as test
variables and English proficiency (TOEFL score) as a grouping variable. Ta-
ble 3 presents the test results and medians of each motivational reason in JH,
HS, and U1 for All Students as well as those in the High Group and the Low
Group. The χ2 ratios were evaluated at p < .05. Statistical significance was
found for five motivational reasons in All Students, for four reasons in the
High Group, and for four reasons in the Low Group. Among them, reasons 3
(short-term goals), 4 (medium-term, specific goals), and 5 (long-term goals)
consistently differed significantly in All Students and the two proficiency
groups. On the contrary, reasons 1 (interest in the target language and cul-
ture), 2 (enjoyment of learning), and 6 (influence of significant others)
did not differ consistently over time in All Students or the two proficiency
groups. Reason 6 (influence of significant others) was particularly stable in
the Low Group (median = 2) and showed no statistically significant differ-
ence. Kendall’s W indicated weak relationships among the variables. These
results suggest that goal-related reasons changed over time, while other
reasons did not; these findings partially support the fourth hypothesis.

Hypothesis 5: The rank orders of motivational reasons between high and

low proficiency university students differ.
This hypothesis was evaluated by calculating Mann-Whitney U tests with
each of the six motivational reasons for each period of schooling as test
variables and the TOEFL score as a grouping variable. Table 4 presents the
results of the tests and the medians of each motivation reason in JH, HS, and
U1 for the High Group and the Low Group. The results of z-approximation
tests showed statistically significant differences in the motivational reasons
between the two proficiency groups when the medians differed by more
than 1. Statistically significant differences were found for motivational rea-
sons 4 (medium-term, specific goals) and 5 (long-term goals) in JH; reasons
2 (enjoyment of learning), 3 (short-term goals), 4 (medium-term, specific
goals), and 5 (long-term goals) in HS; and reason 5 (long-term goals) in U1.
These results suggest that the motivation ranks between the two groups
were the most varied in JH and least varied in U1. Motivational reasons 1
(interest in the target language and culture) and 6 (influence of significant
others) showed no statistically significant differences over time, while the
other reasons showed at least one statistically significant difference. This
44 JALT Journal, 32.1 • May 2010

Table 3. Motivation Ranks

Motivational Median
reason JH HS U1 χ2 Kendall’s W
All Students 1 4 4 5 10.50* .04
(N = 138) 2 4 3 3 7.11* .03
3 5 4 3 51.13* .19
4 3 4 3 19.59* .07
5 3 4 4 22.83* .08
6 2 2 2 4.04 .02
High 1 4 4 5 .36 .00
(N = 83) 2 4 4 3 4.04 .02
3 4 3 3 32.59* .20
4 2 4 3 12.11* .07
5 4 4 5 13.46* .08
6 2 2 2 8.50* .05
Low 1 3 4 5 21.31* .19
(N = 55) 2 4 3 3 3.06 .03
3 5 5 4 20.69* .19
4 4 5 3 13.58* .12
5 2 3 3 9.40* .09
6 2 2 2 .70 .01
Note. The motivational reasons are: 1 and 2 = Intrinsic motivation, 3 = Short-term
goals, 4 = Medium-term goals, 5 = Long-term goals, 6 = Influence of others. Schools
are: JH = junior high school, HS = high school, U1 = the 1st year of university.
*p < .05.

suggests that significant differences mostly emerged among the goal-related

reasons as noted in the results for Hypothesis 4. Among these goal-related
reasons, reasons 3 (short-term goals) and 4 (medium-term, specific goals)
were ranked high in the Low Group, while reason 5 (long-term goals) was
ranked high in the High Group in JH, HS, and U1 (See Table 4). Thus, the
students in the High Group ranked long-term goals higher than short- and
Miura 45

medium-term goals, while those in the Low Group ranked short- and medi-
um-term goals higher than long-term goals. Taken together, these findings
provide strong support for the fifth hypothesis: that high and low proficiency
students differ in their rank orders of motivational reasons.

Table 4. Motivational Rank Differences Between the High and Low

Proficiency Group
Median
Motivational High Low
reason Group Group Mann-Whitney U z
JH 1 4 3 2025.50 -1.14
2 4 4 2128.50 -.69
3 4 5 1860.00 -1.89
4 4 4 1551.50 -3.26*
5 4 2 1714.50 -2.51*
6 2 2 2118.00 -.74
HS 1 4 4 1885.00 -1.76
2 4 3 1781.50 -2.23*
3 3 5 1594.50 -3.07*
4 4 5 1685.00 -2.68*
5 4 3 1702.00 -2.56*
6 2 2 2083.00 -.90
U1 1 5 5 2033.50 -1.11
2 3 3 1909.50 -1.66
3 3 4 1933.00 -1.55
4 3 3 2037.50 -1.08
5 5 3 1586.50 -3.11*
6 2 3 1963.50 -1.44
Note. Motivational reasons and schools are as in Table 3.
*p < .05 (2-tailed).
46 JALT Journal, 32.1 • May 2010

Discussion
The statistical results mostly supported the five a priori hypotheses. Hy-
potheses 1 through 3 concerned the change of motivational levels. Regarding
Hypothesis 1, the results indicate that the students’ L2 learning motivational
levels have frequently changed over time. In previous longitudinal studies,
such as Gardner et al. (2004) and Irie (2005), the researchers reported that
learners’ motivation level was relatively stable during the period under
study; however, when investigated over a longer time period that began with
the commencement of the students’ initial classroom experiences studying
English, their motivational levels clearly displayed frequent changes. One
possible reason for this finding was that the participants in this study were
asked to assess their motivational change over a 7-year period that involved
experiences at three school levels (JH, HS, and U1), with numerous teachers,
and with two high-stakes entrance examinations. In previous longitudinal
studies, by contrast, the researchers investigated students’ motivational
change in one course or in a single educational institution. This difference
has possibly led to the different results.
A second difference from previous findings was that the participants’
motivational levels displayed a general increasing trend throughout the
period under study. This is the opposite of the trend reported in previous
longitudinal studies, in which the learners’ motivational levels gradually
decreased (e.g., Gardner et al., 2004). This difference might be attributable
to the relatively strong motivation of the participants in the present study.
As introduced in the methods section, the university department where this
study was conducted is known to provide a rigorous English program, so the
majority of the participants had to have relatively positive learning histories
in order to be able to enter the program.
Although the participants’ motivational levels displayed a general increas-
ing trend, relatively sharp increases occurred twice between JH2 and JH3,
and between HS2 and HS3, while relatively sharp decreases occurred twice
between JH3 and HS1, and between HS3 and U1. These sharp increases and
decreases, a pattern similar to that observed by Sawyer (2007), occurred
when the entrance examinations for high school and university took place;
these findings indicate the powerful influence that entrance examinations
can exert on Japanese students’ motivation. In Japan, high school and univer-
sity entrance examinations remain unarguably high-stakes tests that largely
determine students’ future courses, a feature in the Japanese educational
landscape that has not changed since Berwick and Ross (1989) conducted
their study two decades ago. Therefore, it is inevitable that in many cases,
Miura 47

motivation for learning English increases before the test and decreases
afterwards. This implies that many Japanese secondary school students
perceive passing entrance examinations, especially university examinations,
as an ultimate future goal and that proximal sub-goals may partly consist of
succeeding on the term-end tests that they take in secondary schools and
mock examinations that they take in supplementary prep schools, instead
of perceiving the entrance examination as a proximal sub-goal for long-dis-
tance goals, such as studying abroad to earn a degree in a foreign university
or working in an international business. This lack of long-term goals may be
one reason why the majority of Japanese university students appear demo-
tivated to learn English and eventually fail to attain high proficiency.
Looking at the change of motivational levels in the two proficiency groups
(Hypothesis 3), both groups were similar in the first 2 years, but started
to differ in their final year of JH. Furthermore, the High Group maintained
a higher degree of motivation than the Low Group throughout JH and U1.
One possible cause of this finding is the different amount of motivational
increase that occurred before the entrance exams and the different amount
of motivational decrease that occurred afterward in the two groups. The
statistical results showed that the increase before the tests was larger in
the High Group than in the Low Group, while the decrease after the tests
was larger in the Low Group than in the High Group. These two changes,
which occurred when the participants were in their final years of junior
and senior high school, might have partly determined their current English
proficiencies. The students who increased their motivational level before
taking the entrance examination were more likely than students with lower
levels of motivation to score well and be satisfied with the test results and
were therefore better able to keep their motivational levels relatively high,
a situation that may have contributed to their higher current proficiency.
This interpretation is consistent with Ushioda’s (1998) finding that the mo-
tivated students in her study perceived their past learning experiences as
the most influential factor affecting their motivation and that successful past
learning experiences generated future motivation.
Hypotheses 4 and 5 concerned changes in the rank order of motivational
reasons in different years in school. Hypothesis 4 was partly supported: the
rank order of the three goal-related reasons consistently changed, while
the intrinsic and social reasons were stable. This finding is almost certainly
related to the ranking of the reasons; the students ranked the goal-oriented
reasons relatively high, intrinsic and social reasons relatively low, and the
social motivational reason was ranked low for all three school periods. On
48 JALT Journal, 32.1 • May 2010

one hand, this result makes sense when considering the major impact of
entrance examinations on motivation suggested in the results for Hypoth-
eses 2 and 3. The students were pressured to become goal oriented because
of the two high-stakes tests that they faced at pivotal learning stages. On
the other hand, this result is unexpected because intrinsic reasons, such as
enjoying learning English, have usually been found to play important mo-
tivational roles in the case of relatively motivated learners (Brown, 2001,
pp. 76-77; Nakata, 2006; Ushioda, 1998). Furthermore, previous studies
have suggested that many Asian English learners receive relatively strong
motivational influences from people close to them such as friends or fam-
ily members (Sawyer, 2007; see also Chen, Warden, & Chang, 2005). The
current result is consistent with Hayashi’s (2005) suggestion that initial
intrinsic motivation is insufficient to sustain long-term motivation, and
that students who sufficiently internalize extrinsic goals, such as passing
entrance examinations, succeed in maintaining high levels of motivation.
The adequate internalization of extrinsic goals may lead to the development
of stronger intrinsic motivation and autonomous learning (Nakata, 2006).
Among the three goal-related reasons, and regarding Hypothesis 5, the
High Group ranked the long-term goal relatively high, while the Low Group
ranked the short-and medium-term goals relatively high. This difference
indicates that the higher proficiency students tended to focus on long-term
goals, such as their future career, while the lower proficiency students tend-
ed to target immediate goals, such as passing the next test or earning cred-
its. Miller and Brickman (2004) argued that learners who seek long-term
goals are likely to set short-term sub-goals that allow them to consequently
achieve their long-term goals. The higher proficiency students in this study
who established long-term goals might have successfully achieved specific
sub-goals (e.g., success on term tests), but the continued presence of long-
term goals motivated them to continue studying and achieve their current
higher proficiency levels.

Conclusion
This study resulted in four main findings:
1. The participants’ L2 learning motivation frequently fluctuated over
the 7 years.
2. High school and university entrance examinations strongly influ-
enced the participants’ motivation.
Miura 49

3.The participants were more goal oriented rather than either intrinsi-
cally or socially motivated.
4. The higher proficiency participants generally perceived entrance ex-
aminations as proximal sub-goals that would help them achieve dis-
tant future goals, while the lower proficiency participants perceived
passing entrance examinations as their ultimate future goal.
These findings imply that having distant future goals that go beyond
passing a university entrance examination is important for sustaining the
long-term learning motivation that leads to higher levels of foreign language
proficiency. As many Japanese students need to keep studying English after
graduating from university if they wish to become highly proficient users
of English, university English courses should be a source of long-term goals
by providing students with meaningful answers to the question of why they
need to study English now and in the future.

Notes
1. The five juniors and two seniors were repeaters who had failed to pass
the courses when they were freshmen.
2. For the motivational rank order section of the questionnaire, the data
from only 138 participants were entered because 56 answered as if they
were responding to a Likert scale instead of rank-ordering the options.

Tsuyuki Miura is a doctoral candidate at Temple University Japan and an

English instructor at several universities in the Kansai area, Japan. Her re-
search interest is in L2 learner motivation.

References
Berwick, R., & Ross, S. (1989). Motivation after matriculation: Are Japanese learners
of English still alive after exam hell? JALT Journal, 11, 193-210.
Brown, H. D. (2001). Teaching by principles: An interactive approach to language
pedagogy. Englewood Cliffs, NJ: Prentice Hall.
Chen, J. F., Warden, C. A., & Chang, H. (2005). Motivators that do not motivate: The
case of Chinese EFL learners and the influence of culture on motivation. TESOL
Quarterly, 39, 609-633.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human
behavior. New York: Plenum.
50 JALT Journal, 32.1 • May 2010

Dörnyei, Z. (2001). Teaching and researching motivation. Harlow, UK: Pearson Educa-
tion.
Gardner, R. C. (1985). Social psychology and second language learning. London: Ed-
ward Arnold.
Gardner, R. C., Masgoret, A.-M., Tennant, J., & Mihic, L. (2004). Integrative motivation:
Changes during a year-long intermediate-level language. Language Learning, 54,
1-34.
Hayashi, H. (2005). Identifying different motivational transitions of Japanese ESL
learners using cluster analysis: Self-determination perspectives. JACET Bulletin,
41, 1-17.
Hidi, S., & Harackiewicz, J. M. (2000). Motivating the academically unmotivated: A
critical issue for the 21st century. Review of Educational Research, 70, 151-179.
Irie, K. (2005). Stability and flexibility of language learning motivation: A multi-
method study of Japanese junior high school students. Unpublished dissertation
manuscript, Temple University Japan.
Locke, E. A., & Latham, G. P. (1994). Goal setting theory. In H. F. O’Neill & M. Drill-
ings (Eds.), Motivation: Theory and research (pp. 13-29). Mahwah, NJ: Lawrence
Erlbaum.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition,
emotion, and motivation. Psychological Review, 98, 224-253.
Miller, R. B., & Brickman, S. J. (2004). A model of future-oriented motivation and self-
regulation. Educational Psychology Review, 16, 9-33.
Nakata, Y. (2006). Motivation and experience in foreign language learning. Bern, Swit-
zerland: Peter Lang.
Pintrich, P. R., & Schunk, D. H. (1996). Motivation in education. Englewood NJ: Pren-
tice Hall.
Sawyer, M. (2007). Motivation to learning foreign language: Where does it come
from, where does it go? Gengo-to-Bunka, 10, 33-42.
Urdan, T. C., & Maehr, M. L. (1994). Beyond a two-goal theory of motivation and
achievement: A case for social goals. Review of Educational Research, 65, 213-243.
Ushioda, E. (1998). Effective motivational thinking: A cognitive theoretical approach
to the study of language learning motivation. In E. Alcón & V. Codina (Eds.), Cur-
rent issues in English language methodology (pp. 77-89). Castelló de la Plana,
Spain: Publicacions de la Universitat Jaume I.
Miura 51

Wentzel, K. R. (1999). Social-motivation processes and interpersonal relationships:

Implications for understanding motivation at school. Journal of Educational Psy-
chology, 91, 76-97.

Appendix
Changes of Learner Motivation
PART I
• Department_______ Year___ Student number _______Name _________________
• Do you have experiences of studying abroad? (yes / no)
• If yes, what age? From age _____ to age_____ for _____ years in _____________

PART II
The purpose of this survey is to investigate English learners’ motivational
changes. Because this is academic research, your responses have absolutely
no relation to your EC course grades. Thank you for your cooperation!
How has your English learning motivation changed since you were a junior
high school student (JHS), high school student (HS), and university student
(i.e., current) (US). Look at the example chart below, mark your answer with
dots, and connect those dots with lines as shown in the chart.
52 JALT Journal, 32.1 • May 2010

PART III
1. The following three questions (A), (B) and (C) will ask about the motiva-
tion at the three different times that you marked in the chart above. Answer
the questions below by thinking of your overall junior high and high school
motivation.

(A) When you were a junior high school student, what was your motiva-
tion to learn English? Read the following six sentences and rank them from
1 (strongest motivation) to 6 (weakest motivation)

() I was interested in English culture or English speaking people.

() I enjoyed learning English.
() I wanted to earn good grades in my English courses.
() I wanted to succeed the high school entrance exams.
() I wanted to study abroad, or have a job using English, or live in Eng-
lish speaking countries in the future.
() I was influenced to study English by people around me such as my
parents/friends/teachers.

If you were motivated for other reasons, please write them here (no rank is
needed).
____________________________________________________________________________________

____________________________________________________________________________________

(B) When you were a high school student, what was your motivation
to learn English? Read the following six sentences and rank them from 1
(strongest motivation) to 6 (weakest motivation).

() I was interested in English culture or the English speaking people.

() I enjoyed learning English.
() I wanted to earn good grades in my English courses or to gain credits.
() I wanted to succeed the university entrance exams.
Miura 53

() I wanted to study abroad, or have a job using English, or live in Eng-

lish speaking countries in the future.
() I was influenced to study English by people around me such as my
parents/friends/teachers.
If you were motivated for other reasons, please write them here (no rank is
needed).
____________________________________________________________________________________

____________________________________________________________________________________

(C) Currently, what is your motivation to learn English? Read the following
six sentences and rank them from 1 (strongest motivation) to 6 (weakest
motivation).

() I am interested in English culture or the English speaking people.

() I enjoy learning English.
() I want to earn good grades in my English courses or to gain credits.
() I need English for getting a job.
() I want to study abroad, or have a job using English, or live in English
speaking countries in the future.
() I am influenced to study English by people around me such as my
parents/friends/teachers.

If you were motivated for other reasons, please write them here (no rank is
needed).
____________________________________________________________________________________

____________________________________________________________________________________
19-22 November 2010

JALT2010
36th Annual International Conference
on Language Teaching and Learning
& Educational Materials Exposition
Creativity: Think
Outside the Box
Nagoya, Japan
<jalt.org/conference>