I have constructed two alternative metrics for a particular phenomenon, and I am trying see whether they are approximately equal across the dataset (that is, whether they might be reliable "stand ins" for each other in a larger, later regression analysis, or if one is consistently different from the other in a meaningful fashion). Both metrics are ratios and feature identical denominators but alternative measurement strategies with respect to the numerator.
I created a scatterplot of the two metrics and ran a regression between them, but I'm curious what might be considered an appropriate cut-off for assessing consistency between metrics that should be approximately equal.
I'm attaching a screenshot of the two way scatterplot in case that helps.
I created a scatterplot of the two metrics and ran a regression between them, but I'm curious what might be considered an appropriate cut-off for assessing consistency between metrics that should be approximately equal.
I'm attaching a screenshot of the two way scatterplot in case that helps.
Attachments
-
86 KB Views: 0