Linked Life Data

18324950

Source:http://linkedlifedata.com/resource/pubmed/id/18324950

Statements in which the resource exists as a subject.
PredicateObject
rdf:type
lifeskim:mentions
pubmed:issue
4
pubmed:dateCreated
2008-3-7
pubmed:abstractText
Linear regression methods try to determine the best linear relationship between data points while correlation coefficients assess the association (as opposed to agreement) between the two methods. Linear regression and correlation play an important part in the interpretation of quantitative method comparison studies. Their major strength is that they are widely known and as a result both are employed in the vast majority of method comparison studies. While previously performed by hand, the availability of statistical packages means that regression analysis is usually performed by software packages including MS Excel, with or without the software programe Analyze-it as well as by other software packages. Such techniques need to be employed in a way that compares the agreement between the two methods examined and more importantly, because we are dealing with individual patients, whether the degree of agreement is clinically acceptable. Despite their use for many years, there is a lot of ignorance about the validity as well as the pros and cons of linear regression and correlation techniques. This review article describes the types of linear regression and regression (parametric and non-parametric methods) and the necessary general and specific requirements. The selection of the type of regression depends on where one has been trained, the tradition of the laboratory and the availability of adequate software.
pubmed:language
eng
pubmed:journal
pubmed:citationSubset
IM
pubmed:status
MEDLINE
pubmed:month
Apr
pubmed:issn
1742-1241
pubmed:author
pubmed:issnType
Electronic
pubmed:volume
62
pubmed:owner
NLM
pubmed:authorsComplete
Y
pubmed:pagination
529-38
pubmed:meshHeading
pubmed:year
2008
pubmed:articleTitle
How to use linear regression and correlation in quantitative method comparison studies.
pubmed:affiliation
Department of Clinical Biochemistry, The Ipswich Hospital, Ipswich, UK. ptwomey@nhs.net
pubmed:publicationType
Journal Article, Review