Skip to main content
. 2006 Jun 14;60(4):397–404. doi: 10.1136/jcp.2005.034447

Table 3 Distribution of the difference between scores (tissue microarray score–whole section score); test of the hypothesis of no systematic difference. The estimated kappa score quantifies the agreement.

Difference (TMA‐ whole section) Nuclear localisation Cytoplasmic localisation Membrane localisation
Oestrogen receptor Progesterone receptor p27 AIB1 COX2 Bcl2 pS2 HER‐2* EGFR† IGF‐IR
−8 1 1
−7 1 1
−6 1 1 1
−5 1 1 1
−4 1 1 1 3
−3 1 3 4 4 1 1
−2 1 2 5 5 2 4 5
−1 5 8 11 10 8 7 11 2 10
0 35 31 24 28 24 39 34 41 58 37
1 9 3 5 9 7 3 7 1 7
2 3 4 1 1 1
3 1
>4 1
Test of agreement
p Value‡ 0.67 0.20 0.001 0.07 0.07 0.007 0.001 NA NA 0.63
Kappa score 0.49 0.48 0.35 0.46 0.43 0.58 0.43 0.88 NA 0.49
95% CI 0.33% to 0.65% 0.34% to 0.62% 0.20% to 0.50% 0.29% to 0.62% 0.26% to 0.60% 0.42% to 0.73% 0.3% to 0.57% 0.74% to 1.00% 0.29 to 0.70
p Value§ <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01

AIB1, oestrogen receptor co‐regulator amplified in breast cancer 1; COX2, cyclooxygenase 2; EGFR, epidermal growth factor receptor; IGF‐IR, insulin‐like growth factor I receptor; NA, not applicable; TMA, tissue microarrays.

*For the marker HER‐2, only 17 of 44 samples have values above zero. Thus the effective sample size for testing the agreement is smaller than for the other markers.

†It is not possible to test the agreement for the marker EGFR, because all samples have the value 0.

‡Wilcoxon's rank sum test of no systematic difference.

§Test of no association.