2008
DOI: 10.1016/j.jclinepi.2007.11.004
|View full text |Cite
|
Sign up to set email alerts
|

Quality assessment of ordinal scale reproducibility: log-linear models provided useful information on scale structure

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

1
7
0

Year Published

2009
2009
2017
2017

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 9 publications
(8 citation statements)
references
References 28 publications
1
7
0
Order By: Relevance
“…In addition, as recently pointed out by Valet et al. , 22 these results could also indicate that the kappa statistic may indicate an almost perfect degree of agreement, even when a major scale defect exists.…”
Section: Discussionsupporting
confidence: 74%
“…In addition, as recently pointed out by Valet et al. , 22 these results could also indicate that the kappa statistic may indicate an almost perfect degree of agreement, even when a major scale defect exists.…”
Section: Discussionsupporting
confidence: 74%
“…Other approaches are log-linear models, which examine the association of cell counts on the level of categories (eg, observers) and models’ interaction patterns and associations 26 46 47. Similar to the latent class models, there are a handful of examples using log-linear models in GI reliability 16 48. Again, STATA provides commands for latent class models (gllamm) and log-linear models (ipf).…”
Section: What Is the Type Of Outcome Variable And What Is The Most Apmentioning
confidence: 99%
“…Evaluation of distinguishability provides valuable information on the structure of a scale . This method adds another dimension to testing the validity of ordinal scales by highlighting more precisely where they can be improved .…”
Section: Discussionmentioning
confidence: 99%
“…This approach permitted us to refine the scales by either pooling adjacent grades or replacing representative photos to improve the accuracy. Indeed, an equal and high DD value between all adjacent categories of the scale would clearly lead to a lower variability within ratings and hence a better agreement between them …”
Section: Discussionmentioning
confidence: 99%