Table 4 |
|||||||||
|
Summary results of convergent and discriminant validity tests |
|||||||||
|
Item-Discriminant Validity |
|||||||||
|
|
|||||||||
|
Item-Internal Consistency |
Range of Correlations4 |
Test success5 |
|||||||
|
|
|
|
|||||||
|
Factors |
N1 |
Factors |
|||||||
|
Range of Correlations2 |
# Success/Total3 |
1 |
2 |
3 |
4 |
Level 1 |
Level 2 |
||
|
|
|||||||||
|
1 |
7 |
0.558–0.762 |
7/7 |
--- |
0.376–0.465 |
0.456–0.549 |
0.293–.0477 |
1/21 |
20/21 |
|
2 |
4 |
0.604–0.708 |
4/4 |
0.410–0.442 |
--- |
0.343–0.442 |
0.406–0.426 |
0/12 |
12/12 |
|
3 |
5 |
0.535–0.684 |
5/5 |
0.406–0.554 |
0.287–0.406 |
--- |
0.362–0.483 |
1/15 |
14/15 |
|
4 |
3 |
0.550–0.700 |
3/3 |
0.436–0.453 |
0.314–0.491 |
0.427–0.504 |
--- |
0/9 |
9/9 |
|
|
|||||||||
|
1 Number of items, 2 Range of correlations between items and hypothesized factor corrected for overlap, 3 Number of correlations exceeding the 0.40 standard/total number of correlations, 4 Range of correlations between items and other factors, 5 Successful discriminant validity tests. Level 1: Item-factor correlation is higher for hypothesized scale than competing scales. Level 2: Item-scale correlation is significantly higher for hypothesized scale than competing scales. The default significance level for comparing two correlations is two standard errors. The standard error of a correlation coefficient is approximately equal to 1 divided by the square root of the sample size. |
|||||||||
|
Paleologou et al. BMC Health Services Research 2006 6:118 doi:10.1186/1472-6963-6-118 |
|||||||||