The type of Table 2 is known as 2×3 contingency table as it consists of 2 columns and 3 rows comprising six cells. The number of actual frequencies is shown in the right column. The number of expected frequencies is illustrated in the left column. The total number of occurrences in each row called the row totals is presented on the righthand side of Table 1. The total number of cases in each column called the column totals is given in the bottom row. The column total is 40 for the United States and 20 for the United Kingdom.
Table 1: Kappa contingency matrix results
Your record codes  
Mother face  Mother body  Away  Row totals  
Installed record codes  Mother face  A
40 
B
0 
C
0 
MF2
40 
Mother body  D
10 
E
1 
F
0 
MB2
11 

Away  G
2 
H
0 
I
7 
A2
9 

Column totals  MF1
52 
MB1
1 
A1
7 
N
60 
Table 2: Results of expected and actual frequencies
Expected frequencies  Actual frequencies  
Cell A  34.66  40 
Cell E  0.18  1 
Cell I  1.05  7 
Total  35.89  48 
Table 3: Kappa Calculation
Variables  Operation  Result 
Sum of Act. frequencies minus Sum of Exp. frequencies  48 – 35.9  12.1 
Column totals (N) minus Sum of Exp. frequencies  60 – 35.9  24.1 
Divide 1 by 2  12.1 / 24.1  0.5 – shows moderate agreement 
Conclusion
Based on these results of the moderate agreement, it is possible to note that part but not all of the association between two measures may be due to one or more other factors. For example, infants who have difficulties reengaging with their mother after ‘stillface’ may also be introvert (that is reserved in expressing their feelings). Part of the babies’ withdrawnness may be due to mothers’ lack of sensitivity expressed during the play, another part may be due to the babies’ naturally peaceful state of mind. If we removed the influence of the infants’ psychological peculiarities on the association between the babies leaning away from the mothers and the mothers’ behavior, then we can assume that the remaining association between these latter two variables is not the direct result of infants’ psychological type.