An investigator invites severely sleep deprived students to participate in a study to determine whether increased sugar consumption can counteract the negative affects of sleep deprivation in a test-taking situation. Participants consume varying amounts of sugar and take a timed test. Sugar consumption amount and test scores are plotted on the scatterplot below. The investigator runs a linear regression model on these data and obtains a regression coefficient (slope) of -0.01, and a p-value of 0.618. How should the investigator interpret these results? (Not real data. Don't try this at home!) Question 15 options: According to the linear regression model, there was a statistically significant negative association between sugar consumption and test score. According to the linear regression model, there was a statistically significant positive association between sugar consumption and test score. Based on the scatterplot, there is a clear linear pattern in these data. The fact that the p-value was greater than 0.05 suggests that the study was underpowered. Based on the scatterplot, there is a clear non-linear pattern in these data. Therefore a linear regression model is not appropriate for these data