2016
DOI: 10.1080/0022250x.2015.1112384
|View full text |Cite
|
Sign up to set email alerts
|

Understanding and interpreting generalized ordered logit models

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

3
465
0
19

Year Published

2017
2017
2024
2024

Publication Types

Select...
9

Relationship

0
9

Authors

Journals

citations
Cited by 680 publications
(487 citation statements)
references
References 17 publications
3
465
0
19
Order By: Relevance
“…Associations between household characteristics and the 4-level food insecurity variable were analysed using a generalised ordered logistic regression model,26 which allows effect sizes to vary for each interval change in the outcome. This model simultaneously estimates odds ratios for three comparisons: (1) the food secure versus all food insecurity categories; (2) people who are marginally food insecure or food secure versus people experiencing moderate and severe food insecurity; (3) people who are not in severe food insecurity versus people who are in severe food insecurity.…”
Section: Methodsmentioning
confidence: 99%
“…Associations between household characteristics and the 4-level food insecurity variable were analysed using a generalised ordered logistic regression model,26 which allows effect sizes to vary for each interval change in the outcome. This model simultaneously estimates odds ratios for three comparisons: (1) the food secure versus all food insecurity categories; (2) people who are marginally food insecure or food secure versus people experiencing moderate and severe food insecurity; (3) people who are not in severe food insecurity versus people who are in severe food insecurity.…”
Section: Methodsmentioning
confidence: 99%
“…When one of the preferred models—the generalized proportional odds/parallel‐lines model—was estimated with the initial six‐category dependent variable, it gave rise to a large number of negative in‐sample predicted values. It is known that this oddity can be circumvented by combining scale categories (Williams ). Hence, we utilized both the stereotype and multinomial logit models to identify which scale categories are nondistinguishable and thus should be combined.…”
Section: Methodsologymentioning
confidence: 99%
“…Second, its purpose is to control for the anchoring effect problem. It is important to note that omitting this type of unobserved heterogeneity from the specification, and depending on its magnitude, could result in a potential bias of the estimates and therefore of the WTAs (Schneider et al, ; Williams, ).…”
Section: Resultsmentioning
confidence: 99%