Table 2

Calculation of the interobserver agreement (kappa coefficient)

Pathologist 2

Total


1+

2+

3+

1.00

Pathologist 1

1+

Count (n)

7

1

0

8

Pathologist 1 (%)

87.5

12.5

0

100

Pathologist 2 (%)

100

11.1

0

44.4

% of Total

38.9

5.6

0

44.4

2+

Count (n)

0

8

0

8

Pathologist 1 (%)

0

100

0

100

Pathologist 2 (%)

0

88

0

44.4

% of Total

0

44

0

44.4

3+

Count (n)

0

0

2

2

Pathologist 1 (%)

0

0

100

100

Pathologist 2 (%)

0

0

100

11.1

% of Total

0

0

11.1

11.1

Total

Count (n)

7

9

2

18

Pathologist 1 (%)

38.9

50

11.1

100

Pathologist 2 (%)

100

100

100

100

% of Total

38.9

50

11.1

100


Symmetric measures


Value

Standard error (a)

Approximate T (b)

Approximate significance


Kappa coefficient

0.906

0.092

4.894

0.00

Number of valid cases (n)

18


a Not assuming the null hypothesis.

b Using the asymptotic standard error assuming the null hypothesis.

Kavanagh et al. BMC Cancer 2009 9:1   doi:10.1186/1471-2407-9-1

Open Data