Table 4

Intraobserver reliability expressed as the weighted kappa statistic for each pair of observers

Weighted kappa statisticMean weighted kappa statistic for category
For pair 1a v 1b (SE)For pair 1a v 2b (SE)For pair 3a v 3b (SE)
Round0.79 (0.0093)0.75 (0.0097)0.86 (0.0101)0.79
Oval0.86 (0.0065)0.86 (0.0066)0.90 (0.0068)0.86
Superior steepening0.99 (0.0055)0.99 (0.0055)0.98 (0.0054)0.99
Inferior steepening0.98 (0.0062)0.97 (0.0061)0.99 (0.0061)0.98
Symmetric bow tie0.96 (0.0069)0.92 (0.0072)0.96 (0.007)0.96
Symmetric bow tie with skewed radial axes 0.99 (0.0051)1.0 (0.0051)0.99 (0.0051)0.99
Asymmetric bow tie with inferior steepening0.97 (0.0057)0.95 (0.0054)0.99 (0.0058)0.97
Asymmetric bow tie with superior steepening0.97 (0.0054)0.98 (0.0055)1.0 (0.0058)0.98
Asymmetric bow tie with skewed radial axes 1.0 (0.0051)1.0 (0.0052)1.0 (0.0051)1.0
Irregular0.94 (0.0057)0.95 (0.0056)0.99 (0.0059)0.95
Median weighted kappa statistic for whole group0.970.950.990.97
  • There were no maps classified by at least two observers in this category. The weighted kappa statistic is shown here to illustrate the low false positive rate for this category but it has not been used in determining the median weighted kappa statistic for the whole group.