Each row represents an object or an individual. The cell entries is the number of raters
that evaluates the object in question to be in the specific category.

Each row must sum up to the same, i.e. the row-sum of the first row!

E.g.

- 5 objects/patient/X-rays to evaluate

- 3 categories

- 4 raters

1,2,1

2,2,0

0,1,3

3,1,0

4,0,0

Here the row-sum of the first row is 4, i.e. **4** raters.

The first object is evaluated to be in category A by 1 rater, in category B by 2 raters
and in category C by 1 rater. Etc ...

All raters agrees the last object to be in category A

-----------------------------------------------------------------

The Siegel data set has 29 objects, 5 categories and 4 raters.

The Fleiss data set has 10 objects, 3 categories and 5 raters.

For each object (row) enter the number of observers that agree in each category (column). Do separate colunms by tabs or commas, rows with new lines.

The row-sums must equal the number of raters of the**first row** and no missing cell entries allowed.

You can copy/paste Excel data

The row-sums must equal the number of raters of the

You can copy/paste Excel data

Literature:

- Statistical Methods for Rates and Proportions 2

- Non Parametric Statistics for the Behavioral Science 2

Categories | ||||
---|---|---|---|---|

O b j c t s |
<more info> | |||

< 0 Less than chance agreement

0.01–0.20 Slight agreement

0.21–0.40 Fair agreement

0.41–0.60 Moderate agreement

0.61–0.80 Substantial agreement

0.81–0.99 Almost perfect agreement

Return to the Interactive Statistics page

Send e-mail to statpages.org@gmail.com