Relative Average Deviation - RAD

[Soaring Crane]

I know that RAD is used to expression precision, and I know the formula for it:

Per mil RAD = (abs. deviation)/mean X (1000) for parts per thousand.

Once you find the relative average deviation--this number in ppt--what is the significance of it? What is it supposed to tell you about the data? What does this number represent??

This is probably an elementary topic, but I can't seem to grasp a concrete definition and how it applies to experimental data.

Thanks.

 

[EnumaElish]

The (abs. deviation)/mean part is similar to the concept of coefficient of variation. The (1000) ppt part is just rescaling (similar to expressing 0.1 as 10%).

 

[tacman]

RAD, or relative average deviation, is a measure of precision in experimental data. It is calculated by taking the absolute deviation (the difference between each data point and the mean) and dividing it by the mean, then multiplying by 1000 to express it in parts per thousand (ppt). This number represents the average difference between each data point and the mean, relative to the mean itself.

The significance of RAD lies in its ability to provide insight into the consistency and accuracy of the data. A lower RAD indicates a higher level of precision, meaning that the data points are closer to the mean and therefore more consistent. On the other hand, a higher RAD suggests a lower level of precision and a greater spread of data points.

In addition, RAD can be used to compare the precision of different sets of data. For example, if two sets of data have the same mean but different RAD values, the set with the lower RAD would be considered more precise.

Overall, RAD is a useful tool in evaluating the reliability and accuracy of experimental data, and can help researchers determine the validity of their results.