Annexe 3 : La base des données issue de notre
enquête.
|
Obs
|
Réforme
|
Métho
|
Subject
|
Admissi
|
Choix SES
|
AGE
|
Choix B
|
Suppre
|
|
1
|
1
|
0
|
0
|
1
|
1
|
18
|
1
|
0
|
|
2
|
1
|
0
|
1
|
1
|
1
|
18
|
1
|
0
|
|
3
|
1
|
0
|
1
|
1
|
1
|
19
|
1
|
0
|
|
4
|
0
|
1
|
1
|
1
|
1
|
19
|
1
|
0
|
|
5
|
0
|
0
|
1
|
1
|
1
|
19
|
1
|
0
|
|
6
|
|
0
|
1
|
1
|
1
|
19
|
1
|
0
|
|
7
|
|
|
|
1
|
1
|
20
|
1
|
0
|
|
8
|
|
0
|
|
1
|
1
|
20
|
1
|
0
|
|
9
|
1
|
0
|
|
1
|
1
|
20
|
1
|
0
|
|
10
|
0
|
1
|
0
|
1
|
1
|
20
|
1
|
0
|
|
11
|
|
0
|
|
1
|
0
|
21
|
1
|
0
|
|
12
|
1
|
0
|
1
|
1
|
0
|
21
|
1
|
0
|
|
13
|
1
|
0
|
1
|
1
|
1
|
21
|
1
|
|
|
14
|
0
|
1
|
1
|
1
|
1
|
21
|
1
|
0
|
|
15
|
1
|
0
|
1
|
1
|
1
|
21
|
1
|
1
|
|
16
|
1
|
0
|
1
|
1
|
1
|
21
|
1
|
0
|
|
17
|
1
|
1
|
1
|
1
|
1
|
21
|
1
|
0
|
|
18
|
1
|
0
|
1
|
1
|
0
|
22
|
1
|
0
|
|
19
|
1
|
0
|
1
|
1
|
0
|
22
|
1
|
0
|
|
20
|
1
|
1
|
1
|
1
|
1
|
22
|
1
|
0
|
|
21
|
1
|
1
|
1
|
1
|
1
|
24
|
1
|
0
|
|
22
|
1
|
1
|
1
|
1
|
1
|
24
|
1
|
0
|
|
23
|
1
|
0
|
1
|
1
|
1
|
24
|
1
|
0
|
|
24
|
1
|
0
|
1
|
1
|
1
|
24
|
1
|
0
|
|
25
|
1
|
0
|
1
|
1
|
1
|
27
|
1
|
0
|
|
26
|
1
|
1
|
1
|
1
|
1
|
27
|
1
|
0
|
|
27
|
1
|
0
|
1
|
1
|
1
|
27
|
1
|
0
|
|
28
|
1
|
0
|
1
|
1
|
1
|
32
|
1
|
0
|
|
29
|
1
|
0
|
1
|
1
|
1
|
32
|
1
|
0
|
|
30
|
1
|
0
|
1
|
1
|
1
|
35
|
1
|
0
|
|
31
|
1
|
1
|
1
|
1
|
1
|
36
|
1
|
0
|
|
32
|
0
|
1
|
1
|
1
|
1
|
36
|
1
|
0
|
Annexe 4 : Les tableaux récapitulatifs des
résultats
Dependent Variable: REFORME
Method: ML - Binary Probit (Quadratic hill climbing) Date:
06/25/12 Time: 16:06
Sample: 1 32
Included observations: 32
Andrews and Hosmer-Lemeshow Goodness-of-Fit Tests Grouping based
upon predicted risk (randomize ties)
|
Quantile of Risk Low High
|
Actual
|
Dep=0
Expect
|
Actual
|
Dep=1
Expect
|
Total Obs
|
H-L Value
|
|
1
|
0.1221
|
0.3695
|
2
|
2.13880
|
1
|
0.86120
|
3
|
0.03138
|
|
2
|
0.3695
|
0.3803
|
2
|
1.88056
|
1
|
1.11944
|
3
|
0.02033
|
|
3
|
0.6509
|
0.6509
|
0
|
1.04721
|
3
|
1.95279
|
3
|
1.60880
|
|
4
|
0.6509
|
0.6509
|
2
|
1.04721
|
1
|
1.95279
|
3
|
1.33175
|
|
5
|
0.6509
|
0.8886
|
1
|
1.15857
|
3
|
2.84143
|
4
|
0.03055
|
|
6
|
0.8886
|
0.8886
|
1
|
0.33408
|
2
|
2.66592
|
3
|
1.49372
|
|
7
|
0.8886
|
0.8886
|
0
|
0.33408
|
3
|
2.66592
|
3
|
0.37595
|
|
8
|
0.8886
|
0.8886
|
0
|
0.33408
|
3
|
2.66592
|
3
|
0.37595
|
|
9
|
0.8886
|
0.8886
|
1
|
0.33408
|
2
|
2.66592
|
3
|
1.49372
|
|
10
|
0.8886
|
0.8939
|
0
|
0.42951
|
4
|
3.57049
|
4
|
0.48118
|
|
|
Total
|
9
|
9.03819
|
23
|
22.9618
|
32
|
7.24332
|
|
H-L Statistic:
|
|
7.2433
|
|
|
Prob. Chi-Sq(8)
|
|
0.5106
|
|
Andrews Statistic:
|
|
18.2838
|
|
|
Prob. Chi-Sq(10)
|
|
0.0504
|
Dependent Variable: REFORME
Method: ML - Binary Probit (Quadratic hill climbing) Date:
06/25/12 Time: 16:06
Sample: 1 32
Included observations: 32
Prediction Evaluation (success cutoff C = 0.5)
Estimated Equation Constant Probability
Dep=0 Dep=1 Total Dep=0 Dep=1 Total
|
P(Dep=1)<=C
|
4
|
2
|
6
|
0
|
0
|
0
|
|
P(Dep=1)>C
|
5
|
21
|
26
|
9
|
23
|
32
|
|
Total
|
9
|
23
|
32
|
9
|
23
|
32
|
|
Correct
|
4
|
21
|
25
|
0
|
23
|
23
|
|
% Correct
|
44.44
|
91.30
|
78.13
|
0.00
|
100.00
|
71.88
|
|
% Incorrect
|
55.56
|
8.70
|
21.88
|
100.00
|
0.00
|
28.13
|
|
Total Gain*
|
44.44
|
-8.70
|
6.25
|
|
|
|
|
Estimated Equation
Dep=0 Dep=1 Total
|
Constant Probability
Dep=0 Dep=1 Total
|
|
E(# of Dep=0)
|
4.03
|
5.01
|
9.04
|
2.53
|
6.47
|
9.00
|
|
E(# of Dep=1)
|
4.97
|
17.99
|
22.96
|
6.47
|
16.53
|
23.00
|
|
Total
|
9.00
|
23.00
|
32.00
|
9.00
|
23.00
|
32.00
|
|
Correct
|
4.03
|
17.99
|
22.02
|
2.53
|
16.53
|
19.06
|
|
% Correct
|
44.76
|
78.22
|
68.81
|
28.13
|
71.88
|
59.57
|
|
% Incorrect
|
55.24
|
21.78
|
31.19
|
71.88
|
28.13
|
40.43
|
|
Total Gain*
|
16.63
|
6.34
|
9.24
|
|
|
|
|
Percent
|
|
|
|
|
|
|
|
Gain**
|
23.14
|
22.55
|
22.85
|
|
|
|
Tabulation of REFORME and METHO and SUBJECT and ADMISSI
|
and CHOIX_SES
Date: 06/25/12 Time: 15:10
Sample: 1 32
Included observations: 32 Tabulation Summary
|
|
|
|
|
Variable
|
Categories
|
|
|
|
REFORME
|
2
|
|
|
|
METHO
|
2
|
|
|
|
SUBJECT
|
2
|
|
|
|
ADMISSI
|
1
|
|
|
|
CHOIX_SES
|
2
|
|
|
|
Product of Categories
|
16
|
|
|
|
Test Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
11
|
13.41425
|
0.2671
|
|
Likelihood Ratio G2
|
11
|
13.60499
|
0.2556
|
WARNING: Expected value is less than 5 in 87.50% of cells (14 of
16).
Table 7: Conditional table for SUBJECT=0, ADMISSI=1,
|
CHOIX_SES=1:
Count
|
0
|
METHO
1
|
Total
|
|
0
|
2
|
1
|
3
|
|
REFORME 1
|
2
|
0
|
2
|
|
Total
|
4
|
1
|
5
|
|
Measures of Association
|
Value
|
|
|
|
Phi Coefficient
|
0.408248
|
|
|
|
Cramer's V
|
0.408248
|
|
|
|
Contingency Coefficient
|
0.377964
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
0.833333
|
0.3613
|
|
Likelihood Ratio G2
|
1
|
1.184939
|
0.2764
|
WARNING: Expected value is less than 5 in 100.00% of cells (4 of
4).
Table 8: Conditional table for SUBJECT=1, ADMISSI=1,
|
CHOIX_SES=1:
Count
|
0
|
METHO
1
|
Total
|
|
0
|
2
|
3
|
5
|
|
REFORME 1
|
12
|
6
|
18
|
|
Total
|
14
|
9
|
23
|
|
Measures of Association
|
Value
|
|
|
|
Phi Coefficient
|
0.225374
|
|
|
|
Cramer's V
|
0.225374
|
|
|
|
Contingency Coefficient
|
0.219860
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
1.168254
|
0.2798
|
|
Likelihood Ratio G2
|
1
|
1.144460
|
0.2847
|
WARNING: Expected value is less than 5 in 50.00% of cells (2 of
4).
Table 9: Conditional table for ADMISSI=1, CHOIX_SES=1:
|
METHO
|
|
|
Count
|
0
|
1
|
Total
|
|
0
|
4
|
4
|
8
|
|
REFORME 1
|
14
|
6
|
20
|
|
Total
|
18
|
10
|
28
|
|
Measures of Association Value
|
|
|
|
Phi Coefficient 0.188562
|
|
|
|
Cramer's V 0.188562
|
|
|
|
Contingency Coefficient 0.185296
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
0.995556
|
0.3184
|
|
Likelihood Ratio G2 1
|
0.973440
|
0.3238
|
|
Note: Expected value is less than 5 in 25.00% of cells (1 of
4).
|
|
|
Table 10: Conditional table for SUBJECT=0, CHOIX_SES=1:
|
|
|
|
METHO
|
|
|
Count
|
0
|
1
|
Total
|
|
0
|
2
|
1
|
3
|
|
REFORME 1
|
2
|
0
|
2
|
|
Total
|
4
|
1
|
5
|
|
Measures of Association Value
|
|
|
|
Phi Coefficient 0.408248
|
|
|
|
Cramer's V 0.408248
|
|
|
|
Contingency Coefficient 0.377964
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
0.833333
|
0.3613
|
|
Likelihood Ratio G2 1
|
1.184939
|
0.2764
|
WARNING: Expected value is less than 5 in 100.00% of cells (4
of
|
4).
Table 11: Conditional table for
Count
|
SUBJECT=1, CHOIX_SES=1:
METHO
0 1
|
Total
|
|
0
|
2
|
3
|
5
|
|
REFORME 1
|
12
|
6
|
18
|
|
Total
|
14
|
9
|
23
|
|
Measures of Association Value
|
|
|
|
Phi Coefficient 0.225374
|
|
|
|
Cramer's V 0.225374
|
|
|
|
Contingency Coefficient 0.219860
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
1.168254
|
0.2798
|
|
Likelihood Ratio G2 1
|
1.144460
|
0.2847
|
|
WARNING: Expected value is less than 5 in 50.00% of cells (2
of
|
|
|
4).
|
|
|
|
Table 12: Conditional table for CHOIX_SES=1:
|
|
|
|
|
METHO
|
|
|
Count
|
0
|
1
|
Total
|
|
0
|
4
|
4
|
8
|
|
REFORME 1
|
14
|
6
|
20
|
|
Total
|
18
|
10
|
28
|
|
Measures of Association Value
|
|
|
|
Phi Coefficient 0.188562
|
|
|
|
Cramer's V 0.188562
|
|
|
|
Contingency Coefficient 0.185296
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
0.995556
|
0.3184
|
|
Likelihood Ratio G2 1
|
0.973440
|
0.3238
|
Note: Expected value is less than 5 in 25.00% of cells (1 of
4).
Table 13: Conditional table for SUBJECT=0, ADMISSI=1:
|
Count
|
0
|
METHO
1
|
Total
|
|
0
|
3
|
1
|
4
|
|
REFORME 1
|
2
|
0
|
2
|
|
Total
|
5
|
1
|
6
|
|
Measures of Association
|
Value
|
|
|
|
Phi Coefficient
|
0.316228
|
|
|
|
Cramer's V
|
0.316228
|
|
|
|
Contingency Coefficient
|
0.301511
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
0.600000
|
0.4386
|
|
Likelihood Ratio G2
|
1
|
0.908053
|
0.3406
|
WARNING: Expected value is less than 5 in 100.00% of cells (4
of
|
4).
Table 14: Conditional table for SUBJECT=1, ADMISSI=1:
METHO
Count 0 1
|
Total
|
|
0
|
2
|
3
|
5
|
|
REFORME 1
|
15
|
6
|
21
|
|
Total
|
17
|
9
|
26
|
|
Measures of Association
|
Value
|
|
|
|
Phi Coefficient
|
0.260360
|
|
|
|
Cramer's V
|
0.260360
|
|
|
|
Contingency Coefficient
|
0.251960
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
1.762465
|
0.1843
|
|
Likelihood Ratio G2
|
1
|
1.684284
|
0.1944
|
WARNING: Expected value is less than 5 in 50.00% of cells (2
of
4).
Table 15: Conditional table for ADMISSI=1:
METHO
0 1 Total
Count
0
REFORME 1
Total
5 4 9
17 6 23
22 10 32
Measures of Association Value
Phi Coefficient 0.178068
Cramer's V 0.178068
Contingency Coefficient 0.175311
Table Statistics df Value Prob
Pearson X2 1 1.014668 0.3138
Likelihood Ratio G2 1 0.981853 0.3217
Note: Expected value is less than 5 in 25.00% of cells (1 of
4).
Table 16: Conditional table for SUBJECT=0:
METHO
0 1 Total
Count
0
REFORME 1
Total
3 1 4
2 0 2
5 1 6
Measures of Association Value
Phi Coefficient 0.316228
Cramer's V 0.316228
Contingency Coefficient 0.301511
Table Statistics df Value Prob
Pearson X2 1 0.600000 0.4386
Likelihood Ratio G2 1 0.908053 0.3406
WARNING: Expected value is less than 5 in 100.00% of cells (4 of
4).
Table 17: Conditional table for SUBJECT=1:
|
Count
|
0
|
METHO
1
|
Total
|
|
0
|
2
|
3
|
5
|
|
REFORME 1
|
15
|
6
|
21
|
|
Total
|
17
|
9
|
26
|
|
Measures of Association Value
|
|
|
|
Phi Coefficient 0.260360
|
|
|
|
Cramer's V 0.260360
|
|
|
|
Contingency Coefficient 0.251960
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
1.762465
|
0.1843
|
|
Likelihood Ratio G2 1
|
1.684284
|
0.1944
|
|
WARNING: Expected value is less than 5 in 50.00% of cells (2
of
|
|
|
4).
|
|
|
|
Table 18: Unconditional table:
|
|
|
|
|
METHO
|
|
|
Count
|
0
|
1
|
Total
|
|
0
|
5
|
4
|
9
|
|
REFORME 1
|
17
|
6
|
23
|
|
Total
|
22
|
10
|
32
|
|
Measures of Association Value
|
|
|
|
Phi Coefficient 0.178068
|
|
|
|
Cramer's V 0.178068
|
|
|
|
Contingency Coefficient 0.175311
|
|
|
|
Table Statistics
|
df
|
Value
|
Prob
|
|
Pearson X2
|
1
|
1.014668
|
0.3138
|
|
Likelihood Ratio G2 1
|
0.981853
|
0.3217
|
Note: Expected value is less than 5 in 25.00% of cells (1 of
4).
| 
