FURUKAWA, Akio
Head of SSS Extensive Reading Study Group
Head of SEG Tadoku
Department
We began an Extensive Reading (ER,
hereafter) Program for Japanese high school students in 2002 at our juku
(private tutoring school) SEG (Scientific Education Group). In 2006, we started
a new ER Program for junior high school students based on four years experience
of our ER program. The purpose of the new program is to achieve the maximum
impact of ER.
Our students received a total of 48 three hour lessons in a year at SEG. The classes met once a week from 5:15 to 8:15 pm. A Japanese teacher conducted ER lessons for 80 minutes while a native English speaking teacher taught other English skills including grammar, writing, and conversation for 80 minutes after 20 minute break. A total of 234 students participated in the new ER program at SEG in April 2008. The students were divided into several classes, whose average number of students was 10 to 12. The table 1 shows the descriptive statistics of their reading amount and reading hours, where N, M,ƒ° and SD stand for the number, the mean, the sum, and the standard deviation, excluding the data of the eighth and ninth grader newcomers who started ER from 2008 academic year.
Table 1 Descriptive Statistics of Participantsf Reading Performance
Materials used in our program were 30,000 books, 2,000 CDs, and 100
DVDs in various levels and genres (Furukawa et al. 2007). In the first six
months, our students listened to CDs individually using portable CD players while
reading books both in class and at home. Since they had
just started learning English and had very limited knowledge of English, we
encouraged students to read books and also listen to accompanied audio CDs at
the same time. After becoming comfortable with reading, the students chose
reading books, listening to CDs or watching DVDs with English subtitles in
class. One third of students read at home, and the others read only in class
after six months.
We report that how our ninth graders in 2008 had raised their
reading level and how much they marked at a test for Japanese high school
students.
Participants in the program started ER on the first day
of the class in March 2006. The first material used were the Oxford Reading
Tree Series, which are very easy picture books written for British children to
learn to read English.


Figure 1 Six in a Bed from Oxford Reading
Tree Series Stage 1
In parallel with learning basic vocabulary and grammar,
our students raised their reading level gradually and finished reading all the
titles of the Oxford Reading Tree Series from Stag1 to Stage 9 in six months.
Next step was to read other series such as Usborne Young Readings, Walker
Stories, Ladybird Tales, Foundations Reading Library, Oxford Bookworms Starter,
Oxford Dominoes Starter, Macmillan Readers Starter, Penguin Readers Starter and
Cambridge English Readers Starter. In April 2008, most of our ninth grader
students who started ER in March 2006, were reading Penguin Readers Level 23,
Oxford Bookworms Stage 23, Macmillan Readers Level 23 and Cambridge English
Readers Level 23 naturally and fluently.
In 2006 academic year, 44 seventh graders participated
the new ER program. 10 left, 7 entered the program during the year, and the
number of the seventh graders in the program was 41 in the end of the 2006
academic year.
In 2007 academic year, 37 out of 44 continued, 15
entered and the number of the eighth graders in the program was 52. 7 left, 6
entered during the year, and the number of the eighth graders in the program
was 51 in the end of the 2007 academic year. At the end of the year, 11 left
and 40 out of 51 continued the program in 2008.
In 2008 academic year, we had 51 ninth graders, 28 out
of 51 students started ER at their
seventh grade, 12 at eighth grade and the other 11at ninth grade.
The table 2 shows the average, the maximum and the
minimum of total amount of words the ninth graders had read by the end of April
2008.
Table 2 Descriptive Statistics of 9^{th} Gradersf Reading
Performance

N

Words per
student 
Words per
student 
Words per
student 
Hours for SSR in
class 
From 7^{th}
grade 
28 
717,756 
1,615,000 
300,000 
138.7 
From 8^{th}
grade 
12 
587,842 
1,020,000 
197,000 
74.7 
From 9^{th}
grade 
11 
65,966 
13,400 
3,000 
10.7 
The table 3 shows the distribution of total amount of
words that our students had read by the end of April in 2008.
Table 3 Distribution of total amount of words 9^{th} Graders
had read
ƒ° of@Words 
From 7^{th} grade 
From 8^{th} grade 
From 9^{th} grade 
1
200,000 
0 
1 
11 
200,001
 400,000 
3 
1 
0 
400,001
 600,000 
10 
3 
0 
600,001
– 800,000 
7 
4 
0 
800,001
– 1,000,000 
4 
2 
0 
1,000,0011,200,000 
1 
1 
0 
1,200,0011,400,000 
2 
0 
0 
1,400,0011,600,000 
0 
0 
0 
1,600,0011,800,000 
‚P 
0 
0 
Our 40 ninth grader students, excluding the newcomers
from 2008 , had read 677,382 words on average and 6 out of 40 students
accomplished onemillion words reading. Three out of these 6 students had not
learned English at their primary schools, whereas 60% of our ninth graders had some
experience of English learning at their primary schools. This suggests that ER
helps the very beginners catch up with the students who had learned English at
primary school.
At the end of January 2008, 49 out of 51 eighth graders
in the program took the Assessment of Communicative English Exam (ACE,
hereafter). Thirtyfour of 49 were the participants form seventh grade and 15
were from eighth grade. The exam is developed by Association for English
Language Proficiency Assessment (ELPA, hereafter) to evaluate English
ability of Japanese high school students.
The table 4 shows the numbers of nationwide exam takers
in Japan, the average scores and the standard derivations of the exam,
according to the data from ELPA (Ikawa, 2008). NW in
the table stands for nationwide in Japan.@

Nof exam takers 
M for Grammar •Vocab (SD) out of 300 
M for Listening (SD) out of 300 
M for Total (SD) out of 900 

3918 
138 (24) 
141 (34) 
158 (28) 
438 (72) 

4480 
151 (29) 
156 (39) 
168 (31) 
475 (86) 

NW 12^{th} graders 
1441 
168 (41) 
176 (54) 
185 (42) 
529 (123) 
SEG ER 8^{th}
graders 
49 
156 (32) 
175 (48) 
196 (38) 
527 (106) 
SEG ER 8^{th}
graders at public schools 
12 
153 (29) 
174 (50) 
196 (37) 
528 (100) 
SEG ER 8^{th}
graders at private
schools 
37 
162i29j 
178(43) 
193 (39) 
523 (103) 
The average Grammar and Vocabulary score of our eighth
graders is higher than that of nationwide tenth graders, but lower than that of
the twelfth graders, whereas listening and reading scores are much higher than
those of tenth graders and even higher than those of twelfth graders.
The onetail ttest (Welchtest using Excel) shows the
probability that the average of total ACE scores of our 49 eighth graders is
higher than that of nationwide 4480 eleventh graders by chance is 0.07% .
Therefore this concludes that our eighth graders reach a higher level in
English than eleventh graders at the level of 0.1% statistical significance.
The ttest also shows that the probability that the average of ACE listening
scores of our eighth graders is higher than that of nationwide eleventh graders
by chance is 2.27%, therefore our eighth graders reach a higher level in
listening English than eleventh graders at the level of 5% statistical significance.
Our students learned English not only at SEG but also at their regular junior high schools, so the differences in scores might be attributed to the effects of total amount of hours students took to learn English. We will show two reasons that suggest this is not the case ant their improved English skills is because of our ER program.
According to a report from Japan educational publisher iKohji et al, 2005j, 23% of high school students in big cities in Japan go to jukus after their regular schools twice a week on average and the total amount of time of top 23% of ACE exam takers is probably almost as same as our students. The average total score of top 23% of nationwide eleventh graders, 1029 out of 4480, is 591 points, whereas the that of our top 23% eighth graders, 11 out of 49, is 684 points. The possibility that our 23% top studentsf scores are higher than those of nationwide eleventh graders by chance is 0.03% , according to onetail ttest.
Students in most private junior
high schools in Japan learn English five to six hours in a week at school,
whereas students in most public or national junior high schools learn English
three hours at school. Considering this, private school students should get
higher marks in English exam. Twelve students out of 49 SEG students who took
ACE exam were from public or national junior high schools, where they learned
English three hours per week, and other 37 students learned English at least
five hours per week at school. The data at table 3 shows that the average of
total scores of our students in public/national schools is 528 points and that
in private schools is 523 points. The probability that the average of total ACE
scores of public/national school eighth graders in our program is equal to that
of our private school eighth graders is 97.9% according to the twotail ttest.
These two possibilities suggest
the differences between the scores of our eighth graders and those of
nationwide Japanese high school students depend not on the hours they learned
English but on the way they learned it. Therefore the results of these ACE
exam strongly suggest that doing ER has helped the eighth
graders reach a higher level in English than eleventh graders.
References:
Furukawa, A.
(2008, April 22). Happy reading! The best way to learn English [Tadoku special].
The Daily Yomiuri, p.20.
Furukawa, A.,
& Ito, S. (2005). Hyakumango Tadoku
Nyumon 100–œŒê‘½“Ç“ü–å [Toward One
Million Words for Beginners]. Tokyo: CosmoPier.
Furukawa, A., Kanda, M., Mayuzumi,
M., Nishizawa, H., Hatanaka, T., Sato, M., & Miyashita, I. (Eds.). (2007). Eigo tadoku kanzen bookguide ‰pŒê‘½“ÇŠ®‘SƒuƒbƒNƒKƒCƒh
[The complete book guide for extensive reading in English] (2nd ed.). Tokyo:
CosmoPier.
ELPA (2003). ACE
tesuto ACEƒeƒXƒg
[The ACE test manual]
Ikawa, E. (2008).
Personal correspondence. May 2, 2008.
Hamajima K. & Takeuchi. K. (2005). Benesse
Kyouiku Kenkyu Kaihasu Senta Shoho Vol. 133 ,
p.2728 ƒxƒlƒbƒZ‹³ˆçŒ¤‹†ŠJ”ƒZƒ“ƒ^[Š•ñ
[Benesse Education Research and
Development Center Report Vol. 133]. Tokyo: Benesse