14
42
12
4
12
3
5
2009()98-2410-H-009-017-MY22009.08.01-2010.7.31
2008NSC 97-2410-H-009 -059 -2008.12.1-2009.7.31
David Easton(1969)
political community
regime
authorities
media malaiseFallows, 1996; Newton 1999; Robinson 1975; 1976; 1997; 2001
virtuous circleAnsolabehere and Iyengar 1995; Iyengar and Kinder 1987; Luskin 2002; Neuman, Just and Crigler 1992; Norris 1999; 2000
panel study
2004813691113999887
1. 2. 3.
1
366
41.3
294
33.1
355
40.0
389
43.9
495
55.8
567
63.9
517
58.3
487
54.9
26
2.9
26
2.9
15
1.7
11
1.2
887
100.0
887
100.0
887
100.0
887
100.0
200420052006
2two-way turnover table
()
()
200
166
0
366
54.6%
45.4%
0.0%
100.0%
92
384
19
495
18.6%
77.6%
3.8%
100.0%
2
17
7
26
7.7%
65.4%
26.9%
100.0%
294
567
28
887
33.1%
63.9%
3.2%
100.0%
FisherFishers exact testp
3two-way turnover table
()
()
201
93
0
294
68.4%
31.6%
0.0%
100.0%
154
402
11
567
27.2%
70.9%
1.9%
100.0%
0
22
4
26
0.0%
84.6%
15.4%
100.0%
355
517
15
887
40.0%
58.3%
1.7%
100.0%
FisherFishers exact testp
4two-way turnover table
()
()
237
118
0
355
66.8%
33.2%
0.0%
100.0%
151
358
8
517
29.2%
69.2%
1.5%
100.0%
1
11
3
15
6.7%
73.3%
20.0%
100.0%
389
487
11
887
43.9%
54.9%
1.2%
100.0%
FisherFishers exact testp.05
Marginal homogeneity (Stuart-Maxwell)>.05
Marginal homogeneity (Bickenboller)>.05Marginal homogeneity (no diagonals)>.05
Contribution to symmetry 2_&_ =4.05p>.05_&_=1.00p>.05_&_ =0.47p>.05
200420052006
513.0%25.8%0.1%38.9%55, 6.2%49, 5.5%65, 7.3%
6Proportional-odds ModelPOM fixed effect
Coef.
SE
odds
0.35
0.14*
1.42
0.23
0.12
1.26
0.01
0.01
1.01
14
-0.15
0.28
0.86
14
-0.01
0.15
0.99
110
0.37
0.12**
1.45
1.13
0.29***
3.08
0.43
0.27
1.53
*
-0.07
0.06
0.93
*
0.04
0.02*
1.04
*
-0.06
0.04
0.94
*
0.07
0.03*
1.07
*
-0.01
0.03
0.99
*
0.03
0.03
1.03
1.34
0.51
5.91
0.53
log likelihood
-2366.494
Wald2
268.79
PseudoR2
.077
3354
* p
7Proportional-odds ModelPOM random effect
Coef.
SE
odds
0.27
0.09**
1.32
0.03
0.15
1.03
-0.28
0.12*
0.75
0.39
0.15**
1.47
0.27
0.12*
1.31
0.00
0.01
1.00
14
-0.14
0.28
0.87
14
0.00
0.16
1.00
110
0.39
0.12**
1.47
1.08
0.29***
2.94
0.37
0.27
1.45
*
-0.08
0.06
0.92
*
0.04
0.02*
1.04
*
-0.06
0.04
0.94
*
0.07
0.03*
1.07
*
-0.01
0.03
0.99
*
0.03
0.03
1.03
1.44
0.51
6.03
0.54
log likelihood
-2356.068
Wald2
279.45
PseudoR2
.081
3354
* p
8Random-intercept Proportional-odds ModelRI-POM random effect
Coef.
SE
odds
0.34
0.12**
1.41
-0.01
0.19
0.99
-0.41
0.16*
0.66
0.52
0.18**
1.68
0.35
0.15*
1.41
0.00
0.02
1.00
14
-0.26
0.33
0.77
14
0.00
0.20
1.00
110
0.42
0.13**
1.52
1.43
0.35***
4.17
0.34
0.33
1.41
*
-0.07
0.07
0.93
*
0.05
0.03
1.05
**
-0.09
0.04*
0.91
**
0.09
0.04*
1.09
**
-0.03
0.04
0.97
**
0.06
0.04
1.06
fixed partthresholds
1.49
0.58*
1.49
7.25
0.62***
7.25
random part
variances and covariance
1.92
log likelihood
-2273.784
Level1
3354
Level2
1368
* p
political predispositions
cross-sectionalpanel study
()
David Easton(1969)
political community
regime
authorities
()
()
()
()
/
(50.2%)
/(49.8%)
(0.0%)
()
(0.0%)
/
(100.0%)
(0.0%)
()
(0.0%)
/(60.7%)
/
(39.3%)
*