3.1 Vocabulary Test 3.1.1 Overall Analysis
The following results were obtained by multiple statistical analysis in order to
explore the connections among the three phases of vocabulary knowledge. Table 2 shows the results of the vocabulary test. A comparison between grades shows that the third graders exhibited a higher number oftotal correct answers, and that the difference was significant (p < .OOI). This dominance of the third graders was shown in each
`meaning' reception, `spelling' reception, and `sound' reception, with significant difference (p < .OOI), as well as in the subcategories of `meaning' production, `sound' production, and `spelling' production.
With regard to receptive knowledge, the number of correct answers for `spelling'
Table 2
Correct Answers in Veca bittary Test
Receptive Productive 3rd grade (R - 165) 2nd grade (n = 182)
Mean SD Mean SD
t-test`Meaning' `Spelling' in L2
in Ll `Sound' in Ll
`Meaning'reception
5.43 6.95 6.19
2.72 2.15
256
3.54 4.86 4.20
280 288
2.91
6.38***
7.69***
726***
`Spelling' `Meaning' in Ll
in L2 `Sound'in Ll
`Spelling'reception
724
6.32 6.78
2.30 2.52 2.45
6.02 5.l2 5.57
2.90 2.67 2.82
434***
431***
4.44***
csound' in L2
`Soun
`Spelling' in L2
`Meaning' in Ll d'reception
4.81 6.32 5.56
259
2.14 2.49
2.32
325
2.78
2.42 2.56
253
9.21***
12.l9***
l1.1l***
Total 6.18 2.55 4.I8 2.98 8.07***
Note . rk' **p <DOI.
reception was the highest and that for `sound' reception was the lowest in both grades, especially in the second grade, with the number ofcorrect answers for `sound' reception significantly low. `Spelling' reception produces `sound' and `meaning' in Ll. It is
reasonable that those two productions in Ll showed higher means than other mixed productions involving Ll and L2. But in the comparison between `meaning' reception and `sound' reception in Ll and in L2, the lower mean for `sound' reception might indicate that participants could not perceive English sounds correctly.
In contrast to receptive knowledge, `sound' production was the highest and
`spelling production' was the lowest in both grades (see Table 3). The high percentage for `sound' production is due to the fact that the participants were required to write
down those sounds on paper in Ll hatakana in this test. Moreover their romoji
knowledge might help them to produce those sounds. The low percentage for `spelling' production would indicate that participants have diff7iculty spelling out L2 words correctly.No significant differences were revealed between reception and production with regard to `meaning'. On the other hand, significant differences were revealed between reception and production with respect to `sound' and `spelling'. These results might indicate that `sound' reception and `spelling' production could be areas of difficulty in
Table 3
Correet Ans}vers ilt Reeeption and Production
3rd grade 2nd grade
R
1Productive
T' l o l
Productive
[lr l 1 Meaning
2 Spelling
3 Sound
Total632 632 632
54.3
48.1
512
69.5 72.4
70.9
61.9 67.8 55.6
51.2 32.5
4L8
35.4
23.2 29.3
48.6
602
54.4
42.e 55.7 27.8
vocabulary learning for students, especially the second graders.
Next, the incorrect answers on the test were analyzed. Five types of answers (A, B, C, D, and E) were possible for each act of `meaning' reception, `spelling' reception,
and `sound' reception (see Figure 5). Diagram no. 1 below shows the production
process that occurs after `meaning' has been received. First, participants grasped a`meaning' in Ll and produced its `sound' correctly, and then produced its `spelling' correctly. This is answer type A. Answer type B is the pattern in which participants failed to produce a correct `spelling' after producing the correct `sound'. Answer type C is the pattern in which participants failed to produce both `sound' and `spelling' correctly. Answer type D is the pattern in which participants failed to produce the correct `sound', but succeeded in producing the correct `spelling'. In this type,
No.1 No.2
--x---+x-- , ... ' '.'" '' ./. .'.'.'.'i.'. '
gg, l,l, ei"""i."'''"''" `"uai''''''}{egrl.tEig s:..l.\ild .-:tw ii k'''''''i'''' Meli.,r, ti-.l
,
.1'/. Lri• .A ,.• :N'x.ll,l,/ '///l)Nx%. ... '
'i''" Spelkn ' Seeljn
No.3 r.
-thi-xgi:ptirpt-r"x-Y--"s""i-k
Sownd C
k-Meanillg
tN.. be A
..
A
": ,, ., s,. ii" '
Spelhn
Figure 5. Five types of answers in three kinds ofreception.
participants might be considered to have directly produced the `spelling' without going via `sound'. Answer type E is the pattern in which participants produced the `sound' and
`spelling' that did not correspond to the meaning. Diagram no. 2 above shows the production process that occurs after `spelling' has been received. All five types of answers are the same as those described for `meaning' reception except that the arrows go in the opposite direction. Diagram no. 3 shows the production process that occurs
`sound' has been received. All five types in no. 3 are different from those in nos. 1 and 2
in which `sound' plays a mediation role in the production of `spelling' or `meaning', and
'
the procedures take one-way routes. In no. 3, the participants first listened to `sound' and produced `meaning' and `spelling' directly. The procedures in no. 3 take two routes, both starting from `sound'.
Table 4 shows the results of the percentages of answers falling into the five answer types for each type of reception. The percentage with both answers correct and the percentage with both answers incorrect were significantly different between the two grades. In the third grade, the percentage with both answers correct (53.20/o) was much higher than the percentage with both answers incorrect (21 .80/o). However, in the second grade the percentage with both answers correct (34.50/o) was lower than the percentage with both answers incorrect (45.20/o). A total of more than 650/o of answers were incorrect among the second graders. This might indicate that the beginners who have been learning English for just one year are still on the way to acquiring the basis of English vocabulary learning.
The differences between grades were also seen in each type of reception. There were no significEmt differences (F (2, 492) = O.26"'S') among the three types of reception (IC:22.80/o, 2C:21.60/o, 3C:21.00/o) in the percentages of the third graders giving both incorrect answers. The second graders, however, showed significant differences (F (2,
543) = 21.85, p < .OOI) among the three types of reception in terms of the percentage giving both answers incorrectly, with the highest percentage of both incorrect answers in `sound' reception (IC:55.40/o) and the lowest in `spelling' reception (2C:34.80/o).
These differences among each type of reception are believed to depend on whether the participants were required to produce the `spelling' or not. The fact that the highest percentage corresponded to `sound' reception might indicate that it was diffTicult for students to receive `sound' correctly. However, as the results of the third graders showed, learning experience can reduce those diffriculties.
Next, the patterns of breakdovvns were categorized. The breakdowns occurred in
Table 4
Five TJIpes ofAnswers
Receptive Productive
ew3dd
Ans O/o -2dd Ans O/o
-
es:.9
>
Meaning
1A:
IB:
1C:
1D:
1E:
Spelling O Spelling Å~
Spelling Å~
Spelling O S elling A
Sound O Sound O
Solmd Å~
Sound Å~
Soimd A
878 268 376 18
le7
53.3 16.3 22.8 l.1
65
609 275 824 35 70
33.6 15.2 45.4 1.9 3.9
Spelling
2A:
2B:
2C:
2D:
2E:
Meaning O
Meaning Å~
Meaning Å~
Meaning O Meanin A
Sound O Sound O
Sound Å~
Sound x
Sound A
IO08 186 355 34 64
612
1l.3 21.6 2.1 3.9
906
190633 25 66
498
1O.4 34.8
1 .4
3.6
x
-bri Sound e
<
53A:
3B:
3C:
3D:
3E:
Spelling O Spelling Å~
Spelling Å~
Spelling O S elling A
Meaning O Meaning O
Meaning Å~
Meaning Å~
Meaning A
746 297 347 47 213
452
18.0 21.0
28
12.9
364 227 1008 58 164
2e.o l2.5
55A 32
9.0
Total A:
C:
Both Both
o
Å~
2632 1078
532
2l.8
1879 2465
34.5
452
No te.
o:
A:
Ans
cerrect amswer, Å~: mcorrect answer,
two productions are correctly cormected but different from reception, : for each type ofreception there were total of165e answers in3rd grade, and l 82O ansvgrers in 2nd grade.
the production of `meaning', `spelling', and `sound'. Table 5 shows the breakdowns in gaining access between `spelling' and `sound' and between `meaning' and `sound'.
Visual reception represents `meaning' reception and `spelling' reception, while auditory reception represents `sound' reception.
The results showed-that the percentages of breakdowns in the second grade were significantly higher than those in third grade. The breakdowns in auditory perception were especially high, especially in second grade. Moreover, in the second grade, the breakdowns between `spelling' and `sound' (henceforth S-S breakdowns) showed no significant difference from the breakdowns between `meaning' and `sound' (henceforth M-S breakdowns) in both visual reception and auditory reception. In the third grade, the breakdowns in visual reception were the same as in second grade, but in the auditory reception the percentage of S-S breakdowns was significantly higher than that of M-S
Tab1e5
Breakdewns in the Production thocess of `Meaning ', `Sound ', and `Spetling ' Breakdowns
w3dd O/. t
A2}g-g!{!gsldadeO/, t
tVisual reception
I between`spelling'and`sound' E between `me aning' and `sound' Auditory reception
M between`spelling'and`sound'
IV between`meaning'aRd`sotmd'
Total reception
V between`spelling'and`sound'
VI between`meanino'and`sound'
21.9 20.9 39.0 23.9 27.6 21.9
O.60
I058***
459***
275 308
67.8 58.5 41.0 40.1
-1.33
l51
O.34
-2.80**•
-4 .62***
-7.39***
t8.39***
-6.79*+ **
•-
897***
No te.
I =IB+2C+2D+2E,
g=lC+1D+lE+2B, M=3B+3C,
IV=-3D+3C ,
V =IB+2C+2D+2E+B3+3C, VI==1C+1D+lE+2B+3D+3C
**p <Dl, **• *p <.OOI.
(see Fi.eq]re 3.1 and Table 3.3),
breakdowns. In short, the second graders have problems with auditory reception, and it is diff7icult for them to produce `meaning' and `spelling' by listening to `sound'. Third graders also have problems with auditory reception, but only in producing `spelling', not in producing `meaning'. Students are believed to have difficulty in listening to and comprehending spoken English, especially in producing `spelling' from `sound' without visual help. On the other hand, the difficulty in producing `meaning' from `sound' without visual help seemed to decline gradually as the students progressed.
3.1.2 Groups Analysis
In order to explore more features of the process of learning vocabulary, six
sample groups were extracted: the upper group, lower group, M-S breakdowns group, S-S breakdowns group, roma]'i interference group, and no roma]'i interference group.
The upper group and lower group consisted of the participants whose vocabulary test scores were in the top 100/o and bottom 100/o of all scores, for their scores were
conspicuously higher or lower than others. The M-S breakdowns group and S-S
breakdowns group consisted of the participants whose breakdowns in the vocabulary test were in the top about 100/o in each, as the same with the upper and lower groups.The romal-i interference group (henceforth romoji group) was defined as including those showing three instances of romoji interference in `sound' production on the vocabulary test. The no romoji interference group (henceforth no romoji group) included those who showed no romal-i interference in `sound' production on the vocabulary test. Regarding the no romal'i group, only the third grade group was sampled, because over half of the participants in second grade showed no romoji interference, and so the sampling was assumed to be unsuitable in the second grade.
All groups showed the general tendency (see Table 3) in reception and in
production. As shown in the analysis of the overall features earlier in this chapter, diff7iculties in `sound' reception and `spelling' production appeared in all groups.
In a comparison of the upper and lower groups (see Table 6), the mean test score ofthe upper group was not very different between the third grade (50.8) and the second grade (47.6), but that in the lower group in the second grade (5.6) was much lower than that in the third grade (15.6). Moreover the percentages were not very different among the three types ofreception and three types ofproduction in the upper groups. But in the lower groups, large differences were seen. The difficulties of `sound' reception and
`spelling' production were revealed, especially for students in the second grade.
In a comparison of the S-S breakdowns group and M-S breakdowns group (see
Table 7 and Table 8), both groups showed the same features as the overall tendency for all breakdowns. In the second grade, the means of both breakdowns and correct answersTable 6
Cerreet Answers in Reeeption and Prodttction by Upper and Lower Groups
3rd grade 2nd grade
Rece tive i
Productive
Total o/6 1
Productive
Total Upper group
1 Meaning 2 Spelling
3 Seund
Total- 82.8 90.9 86.9 85.6 - 89.7 87.7 82.5 76.3 - 79.4 84.1 79.5 90.3
(n = 32, score = 50.8)
- 78.5 89.l 84.7 - 93.8 69.7 60.3
-77.2 69.4 91.5
(n = 34, score == 47.6)
83.8 89.3 65.0
Lower .qroup
1 Meaning 2 Spelling
3 Sound
Total
- l3.i 38.4 25.8 23.1 - 35.9 29.5 35.e IO.6 - 22.8
29.l ll.9 372
(n = 32, score = 15.6)
- 5.7 13.7 l3A - ,l8D
5A O
-9.4 2.9 15.9
(n = 35, score = 5.6)
9.7
15.7
27
'
showed no significant difference between the S-S breakdowns group and the M-S
breakdowns group. In the third grade, the mean of breakdowns showed a significant difference between the two groups, but the mean of the correct answers did not. In acomparison between the grades, the S-S breakdowns groups showed no significant
Table 7
CorreetAnswers in Reeeptien and Prqduction by S-S and M-S Breakdo",ns Groups
Rec tive l
3rd grade Productive
Total g/o l
2nd grade Productive
Total Breakdowns between
l Meaning 2 Spellmg
3 Sound
Total`spelling' and `soimd' (S-S) group
- 30.e 55.3 29.8 45.0 - 55.9 30.6
50.6 22.5 - 26.1 32.3 l3.9 40.3
(n = 37, score = i2.4)
23.2 21.4 22.3
12.4 33.5
- 26.8 6.5
9.5 3Ql
(n = 32, score = l7.3) 23.0 25.e l3.9
Breakdowns between 1 Meaning
2 Spelling
3 Sound
Total`meaning' and `sound' (M-S) group
- 41.0 59.8 34.3 54.1 - 70.5 43.3 59.e 37.1 - 33.8 38.7 25.l 47.6
(n =41, score = 22.3)
22.4 IO.8 l6.6
5.6 9.2
- 332
5.4 21.2
(n = 25, score = 8.6) 7.4 27.8
8.0
Table 8
Breakdowns and CorrectAns"vers b.v S-S and MLS Breakdewns Gromps
-3dd S-S ou M-S ou rw2dd S-S ou M-S ou S-S ou M-S ou
Breakdowns t-test
9.53 2.29*
8.37 9.35
O.32
9.2e O.35 -1.83
correct answers t-test
l731
-i.93
22.27 12.37
1.47
8.64 l.98 5Al *• **
Note. "' p<.05, *+ ts')p <.Ol , *\' x17J<DOi.
differences in either the mean breakdowns or correct answers, but the M-S breakdowns groups showed a significant difference in the mean number of correct answers. These results might indicate that breakdowns between `meaning' and `sound' at the beginning stage are a problem in learning vocabulary.
In a comparison between the romoji group and the no romoji group (see Table 9), as the sampling process that was previously mentioned indicated, the results of the
'
romal'i and no romoji groups might suggest that second graders rarely use the romal'i strategy to produce `sound'; in other words they have not yet reached the stage of using the romoji strategy. But the third graders were able to use the strategy with many instances of interference. The number of instances of interference in the third grade ranged from three to nine (mean 4.2), and that in the second grade ranged from two to four (mean 2.3). Moreover, no significant difference (t = -O.82 "'S') in test scores was
Table 9
Correct Answers in Reception and Produetion by Remaji and No romaji Greups
3rd grade 2nd grade
Rece tive 1
Productive
Total e/6 1
Productive
Tot 1 Roma7'i interference group
IMeaning
2 Spelling 562 3 Sound 60.0 Total 58.3
47.1 66.5 56.8
- 66.5 61.3 41.3 - 50.9 44.2 66.5
-(n = 52, score - 33.8)
49.2 32.8 41.0
292 51.6 - 60.0 17.2
23.2 55.8
(n = 25, score = 24.0) 40.4 54.6 25.0
No romqli interference group
IMeaning - 56.9 68.3
w
2 Spelling 60.3 - 67.6
3 Sound 51.7 62.4
Total 61.4 54.3 67.9
(n = 29, score - 36.7)
62.6 64.0 57.1
observed between the romoji group (33.8) and the no romoji group (36.7). These scores are close to the mean of the third grade score (37.1). Each group included participants in both the upper and lower groups. These results might suggest that romoji interference is not related to students' achievement.
In order to explore more features of the process of learning vocabulary, another six sample groups were derived using the criterion of an unbalanced number of correct answers among the three types of reception (see Table 10). The `meaning' < `spelling' (henceforth Me<Sp) group is the group whose score for `spelling' reception was much
higher than that for `meaning' reception. The `meaning' > `spelling' (henceforth Me>Sp) group is the group showing the opposite tendency. Four other groups, namely the `meaning' < 'sound' (henceforth Me<So) group, `meaning' > `sound' (henceforth Me>So) group, `spelling' < `sound' thenceforth Sp<So) group, and `spelling' > `sound' (henceforth Sp>So) group, were extracted in the same way. In the third grade, all of
Table 10
CorrectAnswers in Rece tien and Preduetion by Unbalaneed Rece tion Gromps
Rece tive l
Productive
Total o/o l
Productlve
Total
1 Meaning 2 Spelling
3 Sound
TotalMe < Sp (n :: 18, score = 32.8)
- 32.8 68.3
-57.8 36.7 63.1 34.7
53.9 43.3 78.3 73.3
- 47.2 66.1
-Me > Sp (n = 16, score = 42.5)
- 78.8 58.8
-71.3 61.9 65.0 70.3
869 82.8 67.5 63.l
- 66.6 77.2
-Me < So (n = 17, score = 33.2) Me > So (n = 20, score = 40.4) 1 Meaning
2 Spelling
3 Sound
Total54.1 72.9 63.5
34.1
55.3 44.7
53.5 62.4
57.9
43.8 58.2 64.1
- 71.5 72.0
-56.0 38.0
64D 54.8
85.5 78.5 81.0 76.5
- 47.0 83.3
-Sp < So (n = l2, score = 36.3)
wS>S(15oe365)
1 Meaning 2 Spelling
3 Sound
Total46.7 so.e
63.3
53.3
63.3 58.3
65.0 55.0
60.0
59.2 50.8 71.7
- 54.7 78.0
-se.o 2s.o 64.0 41.3
70.0 62.3 84.0 81D
- 39.0
77.0
-these groups could be extracted, but in the second grade one group could not be extracted because of the low score for `sound' reception. Sampling the second grade students was abandoned, and only samples from the third grade were explored.
All groups showed almost the same percentages of correct answers in `meaning' production (63-650/o); nevertheless two other types of production showed different
percentages in each group. All groups also showed a tendency for lower `sound'
reception to result in lower `spelling' production, and for higher `sound' reception to result in higher `spelling' production. This tendency might confirm the overall trend of the correct answers in reception and production.With regard to the number of correct answers, the Me>Sp group showed the
highest score, and the Me>So group showed the second-highest score. In contrast, theMe<Sp group showed the lowest score, and the Me<So group showed the
second-lowest score. This might indicate that strong `meaning' reception could be related to high test scores.
3.2 Vocabulary Learning Strategies
The following results were obtained from the items of the questionnaire on
vocabulary learning strategies. In the analysis of the results, all data were re-divided into three groups using a scale from 1 to 5:Section 1. `I never or seldom use this strategy' (scores of 1 and 2, respectively; `never or
seldom true for me').
Section 2. `I sometimes use this strategy' (score of 3; `sometimes true for me').
Section 3. `I always or nearly always use this strategy' (scores of4 and 5, respectively;
`always or nearly always or often true for me').
Table 11 shows the overall results of vocabulary learning strategy use. The
strategy use was more frequent in the third grade than in the second grade, but only social strategies were used more frequently in the second grade than in the third grade.
In both grades self-management strategies were used the most frequently, and social strategies were used the most infrequently. In particular, `romal'i" (Ql5) was always or nearly always used the most by 81e/o ofthe third graders and 680/o ofthe second graders.
`Self-testing' (Q7) was used the second-most by 470/o of the third graders and 360/o of
Table 11
The Use of Vocabularpt Learning Strategies Strategies
Mean
3rd grade (nF165) 2nd grade (n==182)
sD sl o/e s2 o/, s3 o/,
Mean
SD SI O/, S2 O/, s3 o/,t- test
-as Ql
'g Q2
c,) Q3
totai
2.26 O.94 60.6 2.19 1.06 63.6
1.88 1.l5 75.8
2.11 66.7
30.3 8.5
25.5 10.9 12.l 11.5 22.6 IO.32.l7 2.20 2.29 2.22
Ll3
1.l3 l.40
64.8 66.5
615
64.3
242
19.2
l48
19.4
lID
l4.3 23.1
l6.l
O.80 -O.08 -2.97**
- g Q4 .i g Q5 a• g 86, E Q8
totai
2.72 1.44 49.7 3.00 l.33 37.6 2.74 1.30 44.8 3.24 l.38 30.9 2.66 1.l2 41.2
2.87 40.8
19.4 30.9 27.3 35.2
26.7 285 212 47.3 36.4 212
26.2 32.62.79 2.68 2.98 2.77 2.64 2.77
l.43 l .47
l38
1 .49
}28
45.6 51.1
385
50.5 5l.6 47.5
23.6
l92
24.7
l32 242
2l.O 30.8 29.7 36.8 36.3 23.6 31.4
-O.46 2.l6*
-1.69 3.03**
O.18