A regression equation is obtained for a set of data. After examining a scatterplot, the researcher notices a data point that is potentially an influential observation. How could the researcher confirm that this data point is indeed an influential observation? How should the researcher proceed if the data point is found to be an influential observation?

Question

Quizplus · Accepted Answer

The Answer is To confirm that the data point is an influential observation, the researcher should remove that data point and recalculate the regression equation. If the new regression equation differs markedly from the original regression equation, the researcher can conclude that the point is an influential observation. The researcher should then try to determine the reason for the influential observation. If it is due to a measurement or recording error, the researcher should remove the data point from the data set. If it is a legitimate data point; the researcher may need to remove it nonetheless or obtain additional data, so that the regression analysis is not so dependent on one data point. (Explanations will vary.)

A Regression Equation Is Obtained for a Set of Data