Table 2 |
|||||||
|
Version characteristics, distributions of student responses, and comparisons of versions of the scenario. |
|||||||
|
Version A |
Version B |
Version C |
Version D |
A vs. B |
A vs C |
C vs. D |
|
|
|
|||||||
|
P value (H0: absence of disease) |
0.04 |
0.11 |
0.08 |
0.08 |
|||
|
|
|||||||
|
Evidence from p value favours |
Disease |
No disease |
No disease |
No disease |
|||
|
|
|||||||
|
Likelihood ratio (LR) in favour of disease |
0.25 |
4 |
0.25 |
0.125 |
|||
|
|
|||||||
|
Evidence from LR favours |
No disease |
Disease |
No disease |
No disease |
|||
|
|
|||||||
|
Interpretation of result: |
p < 0.001 |
p = 0.34 |
p = 0.020 |
||||
|
For |
21 (7.5) |
169 (69.0) |
9 (5.5) |
5 (3.3) |
|||
|
Against |
122 (43.6) |
11 (4.5) |
83 (50.3) |
56 (36.8) |
|||
|
Neither for nor against |
137 (48.9) |
65 (26.5) |
73 (44.2) |
91 (59.9) |
|||
|
|
|||||||
|
Interpreted the result as for or against disease |
143 (51.1) |
180 (73.5) |
92 (55.8) |
61 (40.1) |
p < 0.001 |
p = 0.34 |
p = 0.005 |
|
|
|||||||
|
Interpretation excluding the undecided |
p < 0.001 |
p = 0.27 |
p = 0.74 |
||||
|
For |
21 (14.7) |
169 (93.9) |
9 (9.8) |
5 (8.2) |
|||
|
Against |
122 (85.3) |
11 (6.1) |
83 (90.2) |
56 (91.8) |
|||
|
|
|||||||
|
Strength of evidence among majority opinion |
(N = 122) |
(N = 169) |
(N = 83) |
(N = 56) |
P < 0.001 linear trend |
p = 0.43 linear trend |
p = 0.58 linear trend |
|
Very strong |
5 (4.1) |
2 (1.2) |
1 (1.2) |
3 (5.4) |
|||
|
Strong |
22 (18.0) |
54 (32.0) |
13 (15.7) |
12 (21.4) |
|||
|
Moderate |
39 (32.0) |
95 (56.2) |
28 (33.7) |
12 (21.4) |
|||
|
Weak |
51 (41.8) |
17 (10.1) |
39 (47.0) |
26 (46.4) |
|||
|
Absent |
5 (4.1) |
1 (0.6) |
2 (2.4) |
3 (5.4) |
|||
|
|
|||||||
|
Perneger and Courvoisier BMC Medical Research Methodology 2010 10:78 doi:10.1186/1471-2288-10-78 |
|||||||
