Purpose of calculating Ta, Tb & R Square in Linearity Study

[Prasanna Venkatesh]

Dear all,

Can you please explain what is the purpose of calculating Ta, Tb & R Square in linearity study & how to conclude the study by seeing the Graphical analysis chart. I could not able to understand this calculation. Please help me.

 

[Miner]

ta = the test statistic for the null hypothesis that the slope = 0.
tb = the test statistic for the null hypothesis that the intercept = 0.
R^2 is the percentage of variation explained by the linearity.

Ideally, both ta and tb are not statistically different from zero, and R^2 is relatively low. If they are statistically significant, then there is a linearity issue with the gage.

Regarding the graphical analysis, the Bias = 0 line should fall within the 95% CI bands for the regression line. If it does not there is a linearity problem with the gage.