# Consider this example of multiple linear regression (attached image). A research

Consider this example of multiple linear regression (attached image). A research study reports a relationship between LDL cholesterol and blood pressure. The study describes the relationship as linear by the least-squares regression model as y hat = 50 + .70(LDL-C) (where y-hat is the predicted value for BP). For example, if the person’s LDL were 100 mg/dL then his BP would be predicted to be 120 mm Hg. The study is repeated by another researcher. You have read studies where the author writes, “there is a linear relationship between x and y after adjusting for age, weight, and gender.“ How is this adjustment made? This is done by multiple regression. Since gender is a binary variable, it has a value of either zero or one. Age and weight are measured as interval variables in years and lbs. Look at the attached image for the researchers’ report.

One of the uses of multiple regression is to identify whether confounders are present in bivariate regression. If the criterion for confounding is a change in the relationship of the variable by more than ten percent, then the bivariate analysis is confounded.

Questions:

1. Is the original study confounded and if so, by which variables?

2. Is there still a linear relationship between LDL-C and blood pressure?

Answer the two questions and the reason why.