Statistical techniques for assessing the agreement between two instruments

Astrua, Milena; Ichim, Daniela; Pennecchi, Francesca; Pisani, Marco

doi:10.1088/0026-1394/44/5/015

Cited by 24 publications

(19 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The choice is dependent on the type of variables being studied (continuous vs. categorical -Ludbrook 2002, Bland & Altman 2007, the presence of a "goldstandard" method for comparison (Astrua et al 2007), the number of replicate measurements per sample and the study design.…”

Section: Brief Theory Of Testing the Agreement Between Methodsmentioning

confidence: 99%

Towards harmonization of forest deposition collectors - case study of comparing collector designs

Žlindra¹,

Eler²,

Clarke³

et al. 2011

iForest

View full text Add to dashboard Cite

In recent years the harmonization of methods in the frame of the International Co-operative Programme on Assessment and Monitoring of Air Pollution Effects on Forests (ICP Forests) operating under the UNECE Convention on Long-range Transboundary Air Pollution (CLRTAP) has been intensified. Among the C-actions of the FutMon project (LIFE07 ENV/D/000218; 2009-2010) the C1-Dep-22(SI) action was established with the goal to harmonize and develop the deposition monitoring procedures and sampling methods. The sampling equipment, spatial design of sampling plots and sampling frequency throughout Europe vary considerably. Therefore a step-by-step approach was made where the harmonized sampling equipment was developed and tested first. The selected collectors were installed at one observation plot of each participating country where measurements of throughfall and bulk deposition were run in parallel with the national collectors for a period of one year. To evaluate the agreement between methods, different statistical analyses were used including Altman-Bland plots, model II regression, and repeated measures ANOVA. Preliminary results from the “Intensive forest ecosystem monitoring plot” plot Brdo in NW Slovenia show a good agreement between national and harmonized bulk (both funnel-type) collectors, while comparison of throughfall measurements indicates systematic bias between harmonized (funnel-type) and national (gutter-type) collectors

show abstract

Section: Brief Theory Of Testing the Agreement Between Methodsmentioning

confidence: 99%

Towards harmonization of forest deposition collectors - case study of comparing collector designs

Žlindra¹,

Eler²,

Clarke³

et al. 2011

iForest

View full text Add to dashboard Cite

show abstract

“…The examples are taken from [8]. Comparison studies were carried out for two angle-measuring instruments.…”

Section: Examplesmentioning

confidence: 99%

“…We useσ x =σ y = 0.1 . Since no information was given in [8] concerning the degrees of freedom associated witĥ σ x andσ y , we assume, for illustration only, infinite degrees of freedom for them. Figure 2 displays these realizations.…”

Section: Examplesmentioning

confidence: 99%

See 1 more Smart Citation

Fiducial approach for assessing agreement between two instruments

Wang

Iyer

2008

Metrologia

View full text Add to dashboard Cite

This paper presents an approach for making inferences about the intercept and slope of a linear regression model when both variables are subject to measurement errors. The approach is based on the principle of fiducial inference. A procedure is presented for computing uncertainty regions for the intercept and slope that can be used to assess agreement between two instruments. Computer codes for performing these calculations, written using open-source software, are listed.

show abstract

“…Equivalence tests with repeated measures do not seem to have been developed as such. Yet, a large body of literature under the general labels of "determining agreement between instruments" or "method comparison studies" describes tests that are applicable in these conditions (see, e.g., Altman & Bland, 1983;Astrua, Ichim, Pennecchi, & Pisani, 2007;Bland & Altman, 1986Cox, 2006;Dunn & Roberts, 1999;Hawkins, 2002;Lin, 1989Lin, , 1992Lin, , 2000Lin, Hedayat, Sinha, & Yang, 2002;van Stralen, Jager, Zoccali, & Dekker, 2008;Wang & Iyer, 2008;Westgard & Hunt, 1973). The context in which these tests were originally developed involved a comparison of two alternative GARCÍA-PÉREZ AND ALCALÁ-QUINTANA instruments (or methods) measuring the same variable and the goal was to assess the agreement (or equivalence) between the measures provided by either instrument or method.…”

mentioning

confidence: 99%

Testing Equivalence with Repeated Measures: Tests of the Difference Model of Two-Alternative Forced-Choice Performance

Garcı́a-Pérez

Alcalá-Quintana

2011

Span. j. psychol.

View full text Add to dashboard Cite

Solving theoretical or empirical issues sometimes involves establishing the equality of two variables with repeated measures. This defies the logic of null hypothesis significance testing, which aims at assessing evidence against the null hypothesis of equality, not for it. In some contexts, equivalence is assessed through regression analysis by testing for zero intercept and unit slope (or simply for unit slope in case that regression is forced through the origin). This paper shows that this approach renders highly inflated Type I error rates under the most common sampling models implied in studies of equivalence. We propose an alternative approach based on omnibus tests of equality of means and variances and in subject-by-subject analyses (where applicable), and we show that these tests have adequate Type I error rates and power. The approach is illustrated with a re-analysis of published data from a signal detection theory experiment with which several hypotheses of equivalence had been tested using only regression analysis. Some further errors and inadequacies of the original analyses are described, and further scrutiny of the data contradict the conclusions raised through inadequate application of regression analyses. Resolver problemas teóricos o empíricos requiere en ocasiones contrastar la equivalencia de dos variables usando medidas repetidas. El mero planteamiento de este objetivo supone un desafío para la lógica subyacente a los métodos de contraste de hipótesis estadísticas, que están diseñados para evaluar la magnitud de la evidencia contraria a la hipótesis nula y de ningún modo permiten evaluar la evidencia a favor de ella. En algunos contextos aplicados se ha abordado el problema utilizando métodos de regresión y contrastando la hipótesis de que la pendiente es 1 y la hipótesis de que la ordenada en el origen es 0 (o simplemente la primera de ellas cuando se fuerza la regresión "por el origen"). Este trabajo muestra que esa estrategia conlleva tasas empíricas de error tipo I muy superiores a las tasas nominales bajo cualquiera de los modelos de muestreo más comúnmente implicados en estudios de equivalencia. Como alternativa, se propone una estrategia basada tanto en pruebas tipo ómnibus que incluyen contrastes de medias y varianzas como en análisis sujeto a sujeto (cuando la situación lo permita). Un estudio de simulación con estas pruebas muestra que la tasa empírica de error tipo I se ajusta a la tasa nominal y que la potencia de los contrastes es adecuada. A modo de ilustración, se aplican estos contrastes para re-analizar los datos de un experimento psicofísico sobre detección de contraste que originalmente sólo fueron analizados mediante regresión por parte de los autores del estudio, pese a que todas las hipótesis consideradas implicaban equivalencia con medidas repetidas. Nuestro reanálisis permite una inspección más minuciosa de los datos que revela contradicciones entre las características empíricas de los datos y las conclusiones extraídas mediante la aplicación inadecuada de métodos de...

show abstract

Statistical techniques for assessing the agreement between two instruments

Cited by 24 publications

References 11 publications

Towards harmonization of forest deposition collectors - case study of comparing collector designs

Towards harmonization of forest deposition collectors - case study of comparing collector designs

Fiducial approach for assessing agreement between two instruments

Testing Equivalence with Repeated Measures: Tests of the Difference Model of Two-Alternative Forced-Choice Performance

Contact Info

Product

Resources

About