Week 2 Discussion 1 & 2

The study I decided to present that has a false correlation caused by a lurking variable is “The more you exercise, the more pounds you will lose”. In this study, there are several lurking variable options available. I am sure you can find one.

This chart reflects as hours of exercise increase, pounds lost will increase as well with linear association.

“Correlation measures the strength between two quantitative variables” (Sharpe et al., 2019, p. 111). There are three criteria required for correlation: the data must be quantitative, linearity condition or strength of the line (how straight the line is but the analyst determines the acceptable level), and report any outliers included in the data (Sharpe et al., 2019).

As discussed by Sharpe et al. (2019), “no matter how strong the association, no matter how large the r, no matter how straight the form, there is no way to conclude from high correlation alone that one variable causes the other” (p. 115). This leads us to a lurking variable or another data point other than x and y that simultaneously affects both variables. In this case, what data point could be introduced that would change the results of increasing exercise will increase pounds lost?

I ran the data through a regression statistical model as well.  See charts below.

The slope of the .14 or for each hour of exercise, .14 lbs is lost. R squared reflects how close the data are to the regression, .98 is almost perfect. This is also known as the coefficient of determination representing how predictable the independent variable is of the dependent variable.