Correlational analysis of ordinal data: from Pearson’s r to Bayesian polychoric correlation

Choi, Jaehwa; Peters, Michelle; Mueller, Ralph O.

doi:10.1007/s12564-010-9096-y

Cited by 100 publications

(77 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this study criterion validity with DASH was moderate (r=0,68), which was lower than in both the original and the French-Canadian versions (r=0,85). Researchers describe some factors such as sample size and characteristics of the sample that may affect the size of r in Pearson analysis (54)(55)(56). In this study correlation might be lower than other researchers have reported because of differences in the characteristics of the sample.…”

Section: Discussionmentioning

confidence: 72%

Cross-cultural adaptation, reliability and validity of the Turkish version of the Upper Limb Functional Index (ULFI)

Tonga

Durutürk

Gabel

et al. 2015

Journal of Hand Therapy

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 72%

Cross-cultural adaptation, reliability and validity of the Turkish version of the Upper Limb Functional Index (ULFI)

Tonga

Durutürk

Gabel

et al. 2015

Journal of Hand Therapy

View full text Add to dashboard Cite

“…invasiveness and time since introduction; Choi et al . ). Finally, the responses to open‐ended questions (e.g.…”

Section: Methodsmentioning

confidence: 97%

A risk‐based inventory of invasive plant species of Queensland, Australia: Regional, ecological and floristic insights

Osunkoya¹,

Froese

Nicol

et al. 2019

Austral Ecology

View full text Add to dashboard Cite

Invasive alien plant species threaten agriculture and biodiversity globally and require ongoing management to minimise impacts. However, the large number of invasive species means that a risk-based approach to prioritisation is needed, taking into account the spatial scale of management decisions and myriad of available information. Here, we developed a risk-based inventory of invasive plants in Queensland, Australia, using both current species distribution/abundance and the severity of their impacts. Our assessment followed a comprehensive data collection process including a scoping of local government pest management plans, herbarium records, the published literature and structured elicitation of expert knowledge during a series of regional stakeholder workshops. From~300 plant species that were identified as established and/or emerging invaders in the State, only one-third were considered by practitioners to pose significant risks across regions to be considered management priorities. We aggregated regional species lists into a statewide priority list and analysed the data set (107 species) for historical, geographical, floristic and ecological patterns. Regions on the mainland eastern seaboard of the State share similar invasive plant communities, suggesting that these regions may form a single management unit, unlike the western/inland and the extreme far north (Torres Strait Islands) regions, which share fewer invasive plant species. Positive correlations were detected between invasiveness and time since introduction for some but not all plant life forms. Stakeholders identified research and management priorities for the invasive plant list, including biological control options, public awareness/education, effective herbicide use, ecology/taxonomy and risk analysis. In the course of the exercise, a statewide invasive plant priority list of high-, medium-and low-impact scores for policy, research and management was compiled. Finally, our approach to invasive plant species prioritisation highlighted that planning and policy documents are not necessarily reflected at the grassroot level in terms of species identity and management priorities.We subjected the regional priority lists to a quantitative risk assessment, with the aim of generating a priority list to inform State government policy and research investments

show abstract

“…All structural equation modeling was conducted using R statistical software with the “lavaan” package (Rosseel 2012). Because the variables (i.e., PICTS-SV items) included in the CFA were on an ordinal scale of measurement, we used polychoric correlations rather than Pearson product-moment correlations (Choi et al 2010; Holgado–Tello et al 2010). Any zero values in the 4×4 contingency tables used to calculate the polychoric correlations were changed to .001 (Savalei 2011; Y. Rosseel, personal communication, September 18, 2014).…”

Section: Resultsmentioning

confidence: 99%

Psychometric Validation of a Simplified Form of the PICTS for Low-Reading Level Populations

Disabato

Folk

Wilson

et al. 2015

J Psychopathol Behav Assess

View full text Add to dashboard Cite

The Psychological Inventory of Criminal Thinking Styles (PICTS) is one of the most widely used measures of criminal thinking. Although the PICTS has adequate psychometric qualities with many general population inmates, the measurement confound of reading ability may decrease its construct validity in low-literacy inmates. To help resolve this confound, we present psychometric evaluation of a simplified version of the PICTS (PICTS-SV) in which item language was simplified but item content was preserved. We first conducted Lexile analyses to confirm the reading level of the PICTS-SV is significantly lower than the original PICTS (i.e., sixth grade versus ninth grade). We then tested a bifactor model to confirm the PICTS-SV contains the same two factors as the original PICTS: proactive and reactive criminal thinking. These PICTS-SV results are commensurate with the factor structure of the original PICTS. Results suggest the PICTS-SV is a valid alternative for assessing criminal thinking in inmates with low reading ability.

show abstract

Correlational analysis of ordinal data: from Pearson’s r to Bayesian polychoric correlation

Cited by 100 publications

References 16 publications

Cross-cultural adaptation, reliability and validity of the Turkish version of the Upper Limb Functional Index (ULFI)

Cross-cultural adaptation, reliability and validity of the Turkish version of the Upper Limb Functional Index (ULFI)

A risk‐based inventory of invasive plant species of Queensland, Australia: Regional, ecological and floristic insights

Psychometric Validation of a Simplified Form of the PICTS for Low-Reading Level Populations

Contact Info

Product

Resources

About