Position Deviation: Is it a Dependent or Independent Variable?

[Paul F. Jackson]

Position deviations in 2d or 3d are computed from the difference between the basic and actual X, Y, Z position of the feature. Therefore the radius or diameter of deviation is dependent on the independent X, Y, & Z variables.

Isn't the position deviation itself also an independent variable with respect to its magnitude from part to part.

Couldn't the deviation in X be a dependent variable with respect to the feature's form, orientation, and the chosen method of finding its center (least squares, Max inscribed, etc.)

 

[Al Dyer]

Ah,,,, a toughy,

Help Atul!!!!!

Atul is good

Al...

 

[Paul F. Jackson]

I think I found the answer by searching Independent Var.. and Depen..V.. on the web but I would still appreciate some confirmation or rebuff from you guys or gals. Thx

It depends! When it is the response in the position equation Dp=2sqrt((X-Xbasic)^2+(Y-Ybasic)^2+(Z-Zbasic)^2) then it is dependent variable. When it is used in an equation to calculate the mean Dp or predict its capability it is the independent variable and the mean or indice are the dependent variables. The Y=f(X) analogy cascades where the independent X can become a dependent Y. Even the coordinate variable Z can be considered a dependent variable if it is the displacement of the centroid of an eight point least squares estimation of the center of the hole.

 

[Tim Folkerts]

As generally used, a quantity is "independent" if you measure it directly, and "dependent" if you calculate it based on other measurements. "Independent" and "dependent" will change depending on what you measure and what you calculate.

For example. If you measure the location of two features (x1 & x2) and can use them to predict the distance between them (d = x2-x1), then x1 & x2 are independent and d is dependent. If, however, you measure one position (x1)and size (d), and can use them to predict the other position (x2 = x1+d), then x1 & d are independent and x2 is dependent.


So in answer to "Couldn't the deviation in X be a dependent variable with respect to the feature's form, orientation, and the chosen method of finding its center (least squares, Max inscribed, etc.)", then yes, if you measure those attributes and use them to calculate a value for X, then X is dependent on them.

(If you measure X anyway, then X becomes another independent variable. Then you instead are seeing if the calculated X and the measured X are correlated.)


Tim