Professional Documents
Culture Documents
1 2
1
98
213,700
9550 1597
(1)
(2)
(3)
(4)
38
Abstract
The purpose of this study is to explore a model for graduate students capability
development in Taiwan. The study population are 213,700 public and private graduate
students for 98-school year in Taiwan, there are 9550 questionnaires totally by using
proportional allocation random sampling sent out, and returned 1597 valid
questionnaires; then, applied descriptive statistics, 1-Way MANOVA, Pearson
correlation, and structural equation model (SEM) statistical methods to analyze
data. Results and findings are summarized as follows:(1) The graduate students
development capability status are medium to high level, they are confident most in the
ability to develop teacher-student interactions and intellectual growth; (2) There is a
significant medium positive correlation between teaching methods and students
development capability, student-teacher interaction and students development
capability, as well as teaching methods and student-teacher interaction; (3) The
graduate students development capability are different on marital status, school
district, university concentration, professional status, work salary, daily work hours,
and children numbers; (4) The model has established, teaching method had
significant positive influence graduate students development capability in intellectual
growth and learning result, while teaching method and student-teacher interaction
are high correlation. This study provides some suggestions based on the results, and
expects to serve as reference points of future research on increasing graduate students
development capability in Taiwan.
39
John Naisbitt & Patricia Aburdence Megatrends 2000
20052005
1986
16 2011 156 25
10 2009
4.78%
2005
21
2009
40
97.14% 97.10% 7
68.19
8 18
Brophy, 1988
2003
E. N. Goplerued 1980
2000
1994
1998
41
1.
2.
3.
4.
42
1997
1999
2005
J. PiagetJ. Bruner
2005
2005
2001
43
1998
2003
2
34
2009
2009
44
Intellectual
Learning Outcome
Working Together
2008
98
98213,700
1597
4
98
45
10%11%25
36001
1
23546
46.9%
10265
20.4%
14342
28.5%
2103
4.2%
50256
100%
4479
1948
2722
401
9550
202 (47%)
88 (20.5%)
123 (28.6%)
18 (4.2%)
430 (100%)
26Leung Kember
200619-23
t
213
-14.655 .000
11.151 .000
.428 .000
3.430 .001
14.553 .000
-10.463 .000
Q8.
7.880 .000
Q9.
-1.345 .000
Q10.
-5.714 .000
2.410 .016
-13.931 .000
Q1.
Q2.
Q3.
Q4.
Q5.
Q6.
Q19.
Q20.
46
Q21.
8.760 .000
Q22.
1.152 .000
Q23.
-1.611 .000
Q24.
9.121 .000
Q7.
2.255 .024
Q15.
1.362 .000
-11.284 .000
Q17.
4.504 .000
Q18.
6.971 .000
5.571 .000
-2.271 .023
4.408 .000
-7.000 .000
-3.228 .001
3.228 .001
Q16.
Q11.
Q12.
Q13.
Q14.
Q25.
Q26.
Ha
H 7a
H 2a
H 3a
H 4a
H 5a
H 6a
1
47
H a1
H a2
H a3
H a4
H a5
H a6
H a7
3 n=1597
Cronbachs .7 .9
Q24
F5
3
9
9
.888
F1
4
4
.818
F2
5
5
.794
F3
6
5
.855
F4
2
3
.854
F5
.942
48
.886
.821
.790
.826
.839
.941
KMO
.9 PAF Promax
5 41.04%
7.46%5.13%
4.21%3.39%
61.23%2007
M=3.98M=3.99
M=3.85
M=3.80M=3.74
M 3 4
3.87
r .55p<.01
r=.42p<.01
49
r=.67, p<.01
r .53p<.01
r=.43p<.01
r=.54, p<.01
r=.69p<.01
r=.59p<.01
r=.46p<.01
r=.67, p<.01
1-Way MANOVA
12
6
7
2005
9
10
50
11
structural equation modeling, SEM
1
2
34
2006
LISREL MLmaximum likelihood
3 10
Kolmogorov-Smirnov Z, K-S Z(p<.05)
4
3.91
4.01
3.99
3.90
3.94
3.83
3.78
3.64
3.82
3.72
3.93
3.92
3.98
.73
.70
.72
.73
.72
.78
.70
.74
.71
.74
.74
.75
.79
-.84
-.79
-.81
-.78
-.82
-.74
-.58
-.30
-.68
-.55
-.78
-.74
-.74
51
1.43
1.21
1.19
1.29
1.27
.97
.94
-.02
1.07
.43
.95
1.18
.77
K-S Z
.202
.224
.191
.238
.191
.187
.202
.159
.228
.195
.245
.211
.231
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
(offending estimate)
2 1
3
5
SEM
standardized
residuals, SRmodification index, MI
2006
2
1
2
3
4
5
6
7
1
2
3
4
5
6
7
8
9
**
.27
.49**
.07
.08
-.07
.05
.87**
.71
.75
.74
.72
.78
.78
.85
.70
.88
t
**
3.11
7.07**
1.22
1.04
-1.11
1.04
15.52**
31.38**
34.71**
33.12**
32.25**
35.62**
34.90**
38.58**
29.73**
38.68**
.085
.067
.054
.077
.065
.049
.056
.023
.022
.022
.022
.022
.022
.022
.024
.023
10
11
12
13
1
2
3
4
5
6
7
8
9
10
11
12
13
*p<.05**p<.01
52
.69
.85
.86
.77
.50
.42
.46
.48
.40
.39
.28
.51
.22
.53
.27
.26
.40
**
24.13
13.63**
14.37**
21.22**
25.13**
23.92**
24.55**
24.85**
23.51**
20.59**
15.63**
22.73**
9.85**
24.13**
13.63**
14.37**
21.22**
.022
.020
.018
.019
.020
.018
.019
.019
.017
.019
.018
.022
.022
.019
.022
.022
.024
.50
.71**
2 .42
.53
.27
.69**
.85**
.27**
.75**
.74**
.72**
.08
.78**
**
.87
.46
.48
.4
0
**
.49
.26
.40
.86**
.77
**
-.07
.05
**
.78
.85**
.39
.28
.07
*p<.05**p<.01
.70**
.88**
.51
.22
SEM
2
2 2
2
1597
6 absolute fit measures
53
GFI
AGFI
SRMR
RMSEA
***
.9
.9
.05
.05.08
ECVI< ECVIIM
417.14
.961
.936
.0317
.064
.306
.114
18.329
NFI
NNFI
IFI
CFI
RFI
.986
.983
.988
.988
.980
.9
.9
.9
.9
.9 .95
PNFI
PGFI
.695
.581
489.138
.5
.5
AIC< SAIC
182.000
29307.600
327.321
AIC< IAIC
>200
ECVI
ECVISM
ECVIIM
AIC
SAIC
IAIC
CN
ECVI< ECVISM
7
t 1.96
SFL .69.85
Jreskog & Srbom (1993).45
SMC (Bollen,
54
1989)SMC 6
60.8% 50.4%
72.3% 60.8%
77.4 49.0%
72.3 47.6%
72.3 59.3%
(construct reliability, CR) Bagozzi and Yi (1988
.6
.6 .748
(convergent validity
average variance extracted , AVE.548 .666 .5
.71
.023
.50
.75
.022
.56
F1
.74
.022
.55
.72
.022
.52
.78
.022
.61
.78
.022
.61
F2
.85
.022
.72
.70
.024
.49
F3
.88
.023
.77
.69
.022
.48
F4
.85
.020
.72
.85
.018
.72
F5
.77
.019
.59
55
CR
AVE
31.38**
34.71**
33.12**
32.25**
35.62**
.86
.55
34.90**
38.58**
.80
.67
29.73**
38.68**
.77
.63
24.13**
13.63**
.75
.60
14.37**
21.22**
.80
.66
N=1597SFL= standardized factor loading
SE = standard error ;
SMC = squared multiple correlations for structural equations
CR = construct
reliabilityAVE =average variance extracted
3.87
r=.671p<.01r=.54p<.01
r=.67p<.01
1
56
10
H a1
H a2
H a3
H a4
H a5
H a6
H a7
57
58
59
1999
2006SEM- LISREL
2007 - SPSS
2009179 134
2003
2009
1998
1997
2001
20052009.8.3
http://epaper.edu.tw/old/highedu/plan.html
20092009 7
20 http://www.edu.tw/statistics/content.aspx?site_content_sn=8956
20056322-12
2003
2000
60
2005
2005
1-10
2008
Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models.
Academic of Marketing Science, 16, 76-94.
Bollen, K. A. (1989). Structural equations with latent variables. New York:
JohnWiley & Sons.
Brophy, J. (1988). Educating teachers about about managing classrooms and students.
Teaching and Teacher Education, 4(1), 1-18.
Jreskog, K. G. & Srbom, D. (1993). Structural equation modeling with the SIMPLIS
command language. Chicago, IL: Scientific Software International.
Naisbitt, J., Aburdence, P. (1991). Megatrends 2000. New York: William Morrow
and Company, Inc.Clark, R. E. (1984). Research on student thought process
during computer-based instruction. Journal of Instructional Development, 7, 2-5.
Leung, D.Y. P., & Kember, D. (2006). The influence of teaching approach and
teacher-student interaction on the development of graduate capabilities.
Structural equation modeling, 13(2), 264286.
61