# A Rater Agreement

The concept of “advisor agreement” is quite simple and, for many years, the reliability of Interraters has been measured as a percentage of match among data collectors. To obtain the measurement agreement, the statistician established a matrix in which the columns represent the different advisors and the lines of the variables for which the raters had collected data (Table 1). The cells in the matrix contained the values captured by the data collectors for each variable. An example of this procedure can be made in Table 1. In this example, there are two advisors (Mark and Susan). They each record their values for variables 1 to 10. To obtain a percentage of approval, the researcher subtracted Susan`s scores from Marks Scores and counted the resulting number of zeroes. Dividing the number of zeros by the number of variables provides a measure of the agreement between advisors. In Table 1, the agreement is 80%. This means that 20% of the data collected in the study is incorrect, because only one of the advisors can be correct if there is disagreement. This statistic is directly interpreted as a percentage of correct data. The value, 1.00 – percent approval can be understood as a percentage of data that is false.

That is, if the approval percentage is 82, 1.00-0.82 – 0.18 and 18% is the amount of data that the search data is wrong. If advisors tend to accept, the differences between the evaluators` observations will be close to zero. If one advisor is generally higher or lower than the other by a consistent amount, the distortion differs from zero. If advisors tend to disagree, but without a consistent model of one assessment above each other, the average will be close to zero. Confidence limits (generally 95%) It is possible to calculate for bias and for each of the limits of the agreement. Fleiss` kappa expressly allows that, although there are a fixed number of advisors (z.B three), different objects can be evaluated by different people In this competition, the judges agreed on 3 out of 5 points. The approval percentage is 3/5 – 60%. n represents the number of observations (not the number of spleens).

Subsequent extensions of the approach included versions that could deal with “under-credits” and ordinal scales. [7] These extensions converge with the intra-class correlation family (ICC), which allows us to estimate reliability for each level of measurement, from the notion (kappa) to the ordinal (or ICC) at the interval (ICC or ordinal kappa) and the ratio (ICC). There are also variations that may consider the agreement by the evaluators on a number of points (for example.B. two people agree on the rates of depression for all points of the same semi-structured interview for a case?) as well as cases of raters x (for example. B how do two or more evaluators agree on whether 30 cases have a diagnosis of depression, yes/no a nominal variable). The field in which you work determines the acceptable level of agreement. If it is a sporting competition, you can accept a 60% agreement to nominate a winner. However, if you look at the data from oncologists who choose to take a treatment, you need a much higher agreement – more than 90%.

In general, more than 75% are considered acceptable in most areas. The basic measure for Inter-Rater`s reliability is a percentage agreement between advisors.