Limits...
The agreement chart.

Bangdiwala SI, Shankar V - BMC Med Res Methodol (2013)

Bottom Line: The agreement charts provide a visual impression that no summary statistic can convey, and summary statistics reduce the information to a single characteristic of the data.The agreement chart should be used to complement the summary kappa or B-statistics, not to replace them.The graphs can be very helpful to researchers as an early step to understand relationships in their data when assessing concordance.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC 27599, USA. kant@unc.edu

ABSTRACT

Background: When assessing the concordance between two methods of measurement of ordinal categorical data, summary measures such as Cohen's (1960) kappa or Bangdiwala's (1985) B-statistic are used. However, a picture conveys more information than a single summary measure.

Methods: We describe how to construct and interpret Bangdiwala's (1985) agreement chart and illustrate its use in visually assessing concordance in several example clinical applications.

Results: The agreement charts provide a visual impression that no summary statistic can convey, and summary statistics reduce the information to a single characteristic of the data. However, the visual impression is personal and subjective, and not usually reproducible from one reader to another.

Conclusions: The agreement chart should be used to complement the summary kappa or B-statistics, not to replace them. The graphs can be very helpful to researchers as an early step to understand relationships in their data when assessing concordance.

Show MeSH

Related in: MedlinePlus

Agreement charts for the comparison of first versus second mearsurements using four different risk classification scales for breast cancer based on mammographic density patterns [Garrido-Estepa et al. (2010)].
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3733724&req=5

Figure 3: Agreement charts for the comparison of first versus second mearsurements using four different risk classification scales for breast cancer based on mammographic density patterns [Garrido-Estepa et al. (2010)].

Mentions: The agreement charts in Figure 3 clearly show that agreement is quite high as the rectangles within the unit square are fairly darkened and the ‘one category away’ discrepancy means the shaded portion fills up the rectangles for BI-RADS and Boyd, but not for Wolfe and Tabár. They also show that there is little ‘drift’ bias between first and second measure, since the ‘path of rectangles’ lies along the diagonal. We do note some differences among the scales that are not reflected in the numerical summaries – the preferences for the lowest ‘low-risk category’ for BI-RADS, while Wolfe and Tabár rarely even use the lowest ‘low-risk category’. The four plots in Figure 3 are quite different visually. Note also that Boyd having 6 categorizations while the other scales have 4 is not a hindrance in producing charts that enable the visual comparisons.


The agreement chart.

Bangdiwala SI, Shankar V - BMC Med Res Methodol (2013)

Agreement charts for the comparison of first versus second mearsurements using four different risk classification scales for breast cancer based on mammographic density patterns [Garrido-Estepa et al. (2010)].
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3733724&req=5

Figure 3: Agreement charts for the comparison of first versus second mearsurements using four different risk classification scales for breast cancer based on mammographic density patterns [Garrido-Estepa et al. (2010)].
Mentions: The agreement charts in Figure 3 clearly show that agreement is quite high as the rectangles within the unit square are fairly darkened and the ‘one category away’ discrepancy means the shaded portion fills up the rectangles for BI-RADS and Boyd, but not for Wolfe and Tabár. They also show that there is little ‘drift’ bias between first and second measure, since the ‘path of rectangles’ lies along the diagonal. We do note some differences among the scales that are not reflected in the numerical summaries – the preferences for the lowest ‘low-risk category’ for BI-RADS, while Wolfe and Tabár rarely even use the lowest ‘low-risk category’. The four plots in Figure 3 are quite different visually. Note also that Boyd having 6 categorizations while the other scales have 4 is not a hindrance in producing charts that enable the visual comparisons.

Bottom Line: The agreement charts provide a visual impression that no summary statistic can convey, and summary statistics reduce the information to a single characteristic of the data.The agreement chart should be used to complement the summary kappa or B-statistics, not to replace them.The graphs can be very helpful to researchers as an early step to understand relationships in their data when assessing concordance.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC 27599, USA. kant@unc.edu

ABSTRACT

Background: When assessing the concordance between two methods of measurement of ordinal categorical data, summary measures such as Cohen's (1960) kappa or Bangdiwala's (1985) B-statistic are used. However, a picture conveys more information than a single summary measure.

Methods: We describe how to construct and interpret Bangdiwala's (1985) agreement chart and illustrate its use in visually assessing concordance in several example clinical applications.

Results: The agreement charts provide a visual impression that no summary statistic can convey, and summary statistics reduce the information to a single characteristic of the data. However, the visual impression is personal and subjective, and not usually reproducible from one reader to another.

Conclusions: The agreement chart should be used to complement the summary kappa or B-statistics, not to replace them. The graphs can be very helpful to researchers as an early step to understand relationships in their data when assessing concordance.

Show MeSH
Related in: MedlinePlus